正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

SLAM algorithms struggle on endoscopy videos - specularity, textureless, & dynamic near-field lighting. We introduce a Near-Field Light Bundle Adjst. loss (NFL-BA): improves performance of SOTA SLAM, e.g. MonoGS (⬆️35% in tracking, ⬆️48% in mapping). See

Roni Sengupta

2,610 subscribers

19,439 次观看 • 1 年前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

2 条评论

Roni Sengupta 的头像

Roni Sengupta1 年前

Our NFL-BA loss relies on the fact that points closer to the camera (depth map) with orientations facing the camera (normal map) reflect more light. NFL-BA loss is minimized during the Tracking & Mapping phase of existing SLAM frameworks, results shown on 3DGS SLAMs.

Virtue of Selfish Investing (VoSI) 的头像

Virtue of Selfish Investing (VoSI)2 年前

Supercharge your profits using our time-tested, optimized, low risk buying strategies including Undercut & Rally, VooDoo low volume signatures, Buyable Gap-Ups and Pocket Pivots. We combine positive fundamentals with technical timing to stack the odds in your favor.

相关视频

🚀 Excited to attend my first #NeurIPS2025 and present our lighting-aware SLAM method, NFL-BA! 🔗 🔦 Why this matters: Traditional Bundle Adjustment (BA) in SLAM assumes static lighting, but many real scenarios—endoscopy, search & rescue, subterranean robotics—use co-located light + camera, creating dynamic, near-field lighting that breaks these assumptions. ✨ What we introduce: Near-Field Lighting Bundle Adjustment Loss (NFL-BA) — a formulation that explicitly models near-field illumination inside the BA objective, allowing SLAM systems to jointly reason about geometry, appearance, and lighting. 📈 Results: Modeling lighting directly leads to ~38% improvement in mapping & tracking across dynamic-lighting sequences in both endoscopy and indoor scenes. 🧩 Plug-and-play: NFL-BA integrates seamlessly into existing neural rendering–based SLAM pipelines, using both implicit (NeRF-style) or explicit (3DGS) scene representations. 🌟 If you’re interested in SLAM, neural rendering, or illumination modeling, come check it out on Friday's poster session, 11-2pm, #4407.

🚀 Excited to attend my first #NeurIPS2025 and present our lighting-aware SLAM method, NFL-BA! 🔗 🔦 Why this matters: Traditional Bundle Adjustment (BA) in SLAM assumes static lighting, but many real scenarios—endoscopy, search & rescue, subterranean robotics—use co-located light + camera, creating dynamic, near-field lighting that breaks these assumptions. ✨ What we introduce: Near-Field Lighting Bundle Adjustment Loss (NFL-BA) — a formulation that explicitly models near-field illumination inside the BA objective, allowing SLAM systems to jointly reason about geometry, appearance, and lighting. 📈 Results: Modeling lighting directly leads to ~38% improvement in mapping & tracking across dynamic-lighting sequences in both endoscopy and indoor scenes. 🧩 Plug-and-play: NFL-BA integrates seamlessly into existing neural rendering–based SLAM pipelines, using both implicit (NeRF-style) or explicit (3DGS) scene representations. 🌟 If you’re interested in SLAM, neural rendering, or illumination modeling, come check it out on Friday's poster session, 11-2pm, #4407.

Roni Sengupta

21,162 次观看 • 9 个月前

SpecNeRF: Gaussian Directional Encoding for Specular Reflections paper page: Neural radiance fields have achieved remarkable performance in modeling the appearance of 3D scenes. However, existing approaches still struggle with the view-dependent appearance of glossy surfaces, especially under complex lighting of indoor environments. Unlike existing methods, which typically assume distant lighting like an environment map, we propose a learnable Gaussian directional encoding to better model the view-dependent effects under near-field lighting conditions. Importantly, our new directional encoding captures the spatially-varying nature of near-field lighting and emulates the behavior of prefiltered environment maps. As a result, it enables the efficient evaluation of preconvolved specular color at any 3D location with varying roughness coefficients. We further introduce a data-driven geometry prior that helps alleviate the shape radiance ambiguity in reflection modeling. We show that our Gaussian directional encoding and geometry prior significantly improve the modeling of challenging specular reflections in neural radiance fields, which helps decompose appearance into more physically meaningful components.

SpecNeRF: Gaussian Directional Encoding for Specular Reflections paper page: Neural radiance fields have achieved remarkable performance in modeling the appearance of 3D scenes. However, existing approaches still struggle with the view-dependent appearance of glossy surfaces, especially under complex lighting of indoor environments. Unlike existing methods, which typically assume distant lighting like an environment map, we propose a learnable Gaussian directional encoding to better model the view-dependent effects under near-field lighting conditions. Importantly, our new directional encoding captures the spatially-varying nature of near-field lighting and emulates the behavior of prefiltered environment maps. As a result, it enables the efficient evaluation of preconvolved specular color at any 3D location with varying roughness coefficients. We further introduce a data-driven geometry prior that helps alleviate the shape radiance ambiguity in reflection modeling. We show that our Gaussian directional encoding and geometry prior significantly improve the modeling of challenging specular reflections in neural radiance fields, which helps decompose appearance into more physically meaningful components.

AK

33,945 次观看 • 2 年前

PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking paper page: introduce PointOdyssey, a large-scale synthetic dataset, and data generation framework, for the training and evaluation of long-term fine-grained tracking algorithms. Our goal is to advance the state-of-the-art by placing emphasis on long videos with naturalistic motion. Toward the goal of naturalism, we animate deformable characters using real-world motion capture data, we build 3D scenes to match the motion capture environments, and we render camera viewpoints using trajectories mined via structure-from-motion on real videos. We create combinatorial diversity by randomizing character appearance, motion profiles, materials, lighting, 3D assets, and atmospheric effects. Our dataset currently includes 104 videos, averaging 2,000 frames long, with orders of magnitude more correspondence annotations than prior work. We show that existing methods can be trained from scratch in our dataset and outperform the published variants. Finally, we introduce modifications to the PIPs point tracking method, greatly widening its temporal receptive field, which improves its performance on PointOdyssey as well as on two real-world benchmarks.

PointOdyssey: A Large-Scale Synthetic Dataset for Long-Term Point Tracking paper page: introduce PointOdyssey, a large-scale synthetic dataset, and data generation framework, for the training and evaluation of long-term fine-grained tracking algorithms. Our goal is to advance the state-of-the-art by placing emphasis on long videos with naturalistic motion. Toward the goal of naturalism, we animate deformable characters using real-world motion capture data, we build 3D scenes to match the motion capture environments, and we render camera viewpoints using trajectories mined via structure-from-motion on real videos. We create combinatorial diversity by randomizing character appearance, motion profiles, materials, lighting, 3D assets, and atmospheric effects. Our dataset currently includes 104 videos, averaging 2,000 frames long, with orders of magnitude more correspondence annotations than prior work. We show that existing methods can be trained from scratch in our dataset and outperform the published variants. Finally, we introduce modifications to the PIPs point tracking method, greatly widening its temporal receptive field, which improves its performance on PointOdyssey as well as on two real-world benchmarks.

AK

122,567 次观看 • 3 年前

Progressively Optimized Local Radiance Fields for Robust View Synthesis paper page: present an algorithm for reconstructing the radiance field of a large-scale scene from a single casually captured video. The task poses two core challenges. First, most existing radiance field reconstruction approaches rely on accurate pre-estimated camera poses from Structure-from-Motion algorithms, which frequently fail on in-the-wild videos. Second, using a single, global radiance field with finite representational capacity does not scale to longer trajectories in an unbounded scene. For handling unknown poses, we jointly estimate the camera poses with radiance field in a progressive manner. We show that progressive optimization significantly improves the robustness of the reconstruction. For handling large unbounded scenes, we dynamically allocate new local radiance fields trained with frames within a temporal window. This further improves robustness (e.g., performs well even under moderate pose drifts) and allows us to scale to large scenes. Our extensive evaluation on the Tanks and Temples dataset and our collected outdoor dataset, Static Hikes, show that our approach compares favorably with the state-of-the-art.

Progressively Optimized Local Radiance Fields for Robust View Synthesis paper page: present an algorithm for reconstructing the radiance field of a large-scale scene from a single casually captured video. The task poses two core challenges. First, most existing radiance field reconstruction approaches rely on accurate pre-estimated camera poses from Structure-from-Motion algorithms, which frequently fail on in-the-wild videos. Second, using a single, global radiance field with finite representational capacity does not scale to longer trajectories in an unbounded scene. For handling unknown poses, we jointly estimate the camera poses with radiance field in a progressive manner. We show that progressive optimization significantly improves the robustness of the reconstruction. For handling large unbounded scenes, we dynamically allocate new local radiance fields trained with frames within a temporal window. This further improves robustness (e.g., performs well even under moderate pose drifts) and allows us to scale to large scenes. Our extensive evaluation on the Tanks and Temples dataset and our collected outdoor dataset, Static Hikes, show that our approach compares favorably with the state-of-the-art.

AK

140,624 次观看 • 3 年前

CoDeF: Content Deformation Fields for Temporally Consistent Video Processing abs: paper page: present the content deformation field CoDeF as a new type of video representation, which consists of a canonical content field aggregating the static contents in the entire video and a temporal deformation field recording the transformations from the canonical image (i.e., rendered from the canonical content field) to each individual frame along the time axis.Given a target video, these two fields are jointly optimized to reconstruct it through a carefully tailored rendering pipeline.We advisedly introduce some regularizations into the optimization process, urging the canonical content field to inherit semantics (e.g., the object shape) from the video.With such a design, CoDeF naturally supports lifting image algorithms for video processing, in the sense that one can apply an image algorithm to the canonical image and effortlessly propagate the outcomes to the entire video with the aid of the temporal deformation field.We experimentally show that CoDeF is able to lift image-to-image translation to video-to-video translation and lift keypoint detection to keypoint tracking without any training.More importantly, thanks to our lifting strategy that deploys the algorithms on only one image, we achieve superior cross-frame consistency in processed videos compared to existing video-to-video translation approaches, and even manage to track non-rigid objects like water and smog.

CoDeF: Content Deformation Fields for Temporally Consistent Video Processing abs: paper page: present the content deformation field CoDeF as a new type of video representation, which consists of a canonical content field aggregating the static contents in the entire video and a temporal deformation field recording the transformations from the canonical image (i.e., rendered from the canonical content field) to each individual frame along the time axis.Given a target video, these two fields are jointly optimized to reconstruct it through a carefully tailored rendering pipeline.We advisedly introduce some regularizations into the optimization process, urging the canonical content field to inherit semantics (e.g., the object shape) from the video.With such a design, CoDeF naturally supports lifting image algorithms for video processing, in the sense that one can apply an image algorithm to the canonical image and effortlessly propagate the outcomes to the entire video with the aid of the temporal deformation field.We experimentally show that CoDeF is able to lift image-to-image translation to video-to-video translation and lift keypoint detection to keypoint tracking without any training.More importantly, thanks to our lifting strategy that deploys the algorithms on only one image, we achieve superior cross-frame consistency in processed videos compared to existing video-to-video translation approaches, and even manage to track non-rigid objects like water and smog.

AK

153,305 次观看 • 3 年前

SUPPLEMENTING YOUR WORK ENVIRONMENT WITH AN INCANDESCENT LIGHT BULB IMPROVES MITOCHONDRIAL FUNCTION! - A brand new (Jan 23th, 2026) paper published by Glen Jeffery and Edward Barrett demonstrates how adding a 60 W incandescent lamp to participants work desk dramatically improves color contrast, or a proxy for color vision. The intervention here is simply placing an incandescent lightbulb on the desk without direct irradiation on participants. And this is huge: “we have obtained significant balanced improvements in both the protan and tritan range. Previously, exposure to restricted experimental 670 nm resulted in improvements biased strongly in favour of only tritan function. Hence, exposure to full spectrum lighting results in a balanced pattern of improvement in visual performance.” The mechanism seems to be mitochondrial in nature. Short wavelength light, which is dominated in standard LED lighting, reduces mitochondrial function. There’s a reduction in mitochondria ATP production, and therefore a reduced demand for glucose. This intern can increase body weight and serum cytokinin levels. Also, with short wavelength blue light, there is an increase probability of cell aging and death. “It is suggested that this is partly due to 420–450 nm light, dominant in LEDs, being absorbed by porphyrin and the subsequent production of oxygen singlets driving inflammation.” “Conversely, exposure to longer wavelengths is associated with increased mitochondrial membrane potential and increased concentration of mitochondrial complex proteins that have declined with ageing and disease. This in turn is associated with elevated ATP, reduced inflammation and extended average lifespan.“ So, for the love of mitochondria, get long-wavelength light! This can be done in many ways: -sunlight/daylight exposure (does not need to be direct sunlight) -fire light -incandescent light bulbs -LEDs that emit long-wavelength light (yes, LEDs CAN emit red and near-infrared, but most don’t)

Jonathan Jarecki

65,208 次观看 • 6 个月前

MaterialFusion Enhancing Inverse Rendering with Material Diffusion Priors discuss: Recent works in inverse rendering have shown promise in using multi-view images of an object to recover shape, albedo, and materials. However, the recovered components often fail to render accurately under new lighting conditions due to the intrinsic challenge of disentangling albedo and material properties from input images. To address this challenge, we introduce MaterialFusion, an enhanced conventional 3D inverse rendering pipeline that incorporates a 2D prior on texture and material properties. We present StableMaterial, a 2D diffusion model prior that refines multi-lit data to estimate the most likely albedo and material from given input appearances. This model is trained on albedo, material, and relit image data derived from a curated dataset of approximately ~12K artist-designed synthetic Blender objects called BlenderVault. we incorporate this diffusion prior with an inverse rendering framework where we use score distillation sampling (SDS) to guide the optimization of the albedo and materials, improving relighting performance in comparison with previous work. We validate MaterialFusion's relighting performance on 4 datasets of synthetic and real objects under diverse illumination conditions, showing our diffusion-aided approach significantly improves the appearance of reconstructed objects under novel lighting conditions. We intend to publicly release our BlenderVault dataset to support further research in this field.

MaterialFusion Enhancing Inverse Rendering with Material Diffusion Priors discuss: Recent works in inverse rendering have shown promise in using multi-view images of an object to recover shape, albedo, and materials. However, the recovered components often fail to render accurately under new lighting conditions due to the intrinsic challenge of disentangling albedo and material properties from input images. To address this challenge, we introduce MaterialFusion, an enhanced conventional 3D inverse rendering pipeline that incorporates a 2D prior on texture and material properties. We present StableMaterial, a 2D diffusion model prior that refines multi-lit data to estimate the most likely albedo and material from given input appearances. This model is trained on albedo, material, and relit image data derived from a curated dataset of approximately ~12K artist-designed synthetic Blender objects called BlenderVault. we incorporate this diffusion prior with an inverse rendering framework where we use score distillation sampling (SDS) to guide the optimization of the albedo and materials, improving relighting performance in comparison with previous work. We validate MaterialFusion's relighting performance on 4 datasets of synthetic and real objects under diverse illumination conditions, showing our diffusion-aided approach significantly improves the appearance of reconstructed objects under novel lighting conditions. We intend to publicly release our BlenderVault dataset to support further research in this field.

AK

22,959 次观看 • 1 年前

🚀 New Paper: Pixel Reasoner 🧠🖼️ How can Vision-Language Models (VLMs) perform chain-of-thought reasoning within the image itself? We introduce Pixel Reasoner, the first open-source framework that enables VLMs to “think in pixel space” through curiosity-driven reinforcement learning. Current VLMs reason only in text — even when grounded in rich images or videos, their logical steps are verbalized in natural language. This restricts their ability to interrogate visual evidence and demonstrate how conclusions are drawn. 🔍 So we ask: What if we could make VLMs "show their work" by reasoning directly in the pixel space? Inspired by GPT-o3’s "think-in-image" ability, we propose a framework where VLMs use interactive visual operations — zoom, select-frame, highlight — to reason through complex visual inputs. To do this, we design a two-stage training process: Instruction tuning with synthesized visual reasoning traces. Reinforcement learning with curiosity-driven reward to balance exploration between pixel and text reasoning ✨ With this, Pixel Reasoner achieves near-SoTA performance on many information-rich multimodal benchmarks: 📊 84% on InfographicsVQA 🧠 84% on V* benchmark 🧩 74% on TallyQA-Complex It also achieves strong accuracy of 68% on MVBench (a video benchmark). Website: Paper: Code: Demo: (coming soon)

🚀 New Paper: Pixel Reasoner 🧠🖼️ How can Vision-Language Models (VLMs) perform chain-of-thought reasoning within the image itself? We introduce Pixel Reasoner, the first open-source framework that enables VLMs to “think in pixel space” through curiosity-driven reinforcement learning. Current VLMs reason only in text — even when grounded in rich images or videos, their logical steps are verbalized in natural language. This restricts their ability to interrogate visual evidence and demonstrate how conclusions are drawn. 🔍 So we ask: What if we could make VLMs "show their work" by reasoning directly in the pixel space? Inspired by GPT-o3’s "think-in-image" ability, we propose a framework where VLMs use interactive visual operations — zoom, select-frame, highlight — to reason through complex visual inputs. To do this, we design a two-stage training process: Instruction tuning with synthesized visual reasoning traces. Reinforcement learning with curiosity-driven reward to balance exploration between pixel and text reasoning ✨ With this, Pixel Reasoner achieves near-SoTA performance on many information-rich multimodal benchmarks: 📊 84% on InfographicsVQA 🧠 84% on V* benchmark 🧩 74% on TallyQA-Complex It also achieves strong accuracy of 68% on MVBench (a video benchmark). Website: Paper: Code: Demo: (coming soon)

Wenhu Chen

82,829 次观看 • 1 年前

George Knapp: "You ever see any video where we're engaging with them, either we're chasing them, our warplanes are going after a disk, a saucer, an orb or something like that?" Matthew Brown: "So I've seen videos and read reports of engagements, but we're not shooting them... engagements [yes], where we are trying to intercept something... we're up there, protecting our airspace, and interacting with them [UAP] in a way that we would consider hostile." And near the end, Jeremy Corbell references the Yemen UAP incident, indirectly: "What happened was the missile, quote, bounced off... what does that imply, a field around it? I don't know." As they say, hindsight is 20/20, including on descriptors for shoot-downs versus engagements. Reading between the lines here. And in Corbell, Knapp and Brown I trust. #ufo #uap #ovni

George Knapp: "You ever see any video where we're engaging with them, either we're chasing them, our warplanes are going after a disk, a saucer, an orb or something like that?" Matthew Brown: "So I've seen videos and read reports of engagements, but we're not shooting them... engagements [yes], where we are trying to intercept something... we're up there, protecting our airspace, and interacting with them [UAP] in a way that we would consider hostile." And near the end, Jeremy Corbell references the Yemen UAP incident, indirectly: "What happened was the missile, quote, bounced off... what does that imply, a field around it? I don't know." As they say, hindsight is 20/20, including on descriptors for shoot-downs versus engagements. Reading between the lines here. And in Corbell, Knapp and Brown I trust. #ufo #uap #ovni

UAP Juan

86,190 次观看 • 1 年前

RUN BEFORE THE FLOOD TAKES EVERYTHING One Woman. One Cow. One Chance to Escape. Made with seedance 2.5 on NadouPro Prompt: Create a photorealistic, high-budget cinematic action scene in a crowded traditional Asian wet market during heavy rain and severe flooding. A young Asian woman with long wet dark hair, wearing a pink waterproof rain suit and black rubber boots, runs through flooded market aisles while a massive wave of water rushes behind her. Show seafood and vegetable stalls, floating crates, overturned objects, panicked crowds, rushing dirty water, reflections, splashes and realistic wet details. Use fast handheld tracking, close-ups, low-angle water shots, side tracking and wide shots, with realistic physics, motion blur and dynamic camera shake. She jumps over obstacles and runs across raised market surfaces. Near the end, she reaches a raised area where a frightened cow is trapped by the flood. She looks at it and urgently says: “Come on, girl! Move!” She guides the cow toward safety. Keep her face, hairstyle and pink outfit consistent. Dark rainy atmosphere, cool blue-gray tones mixed with warm fluorescent lights, cinematic depth of field, volumetric lighting, realistic skin, water and reflections, subtle film grain, 4K anamorphic Hollywood action style. Add large white brush-stroke typography “RUN” briefly across the center. No cartoon look, CGI appearance, distorted anatomy, extra fingers, duplicated people or face changes.

RUN BEFORE THE FLOOD TAKES EVERYTHING One Woman. One Cow. One Chance to Escape. Made with seedance 2.5 on NadouPro Prompt: Create a photorealistic, high-budget cinematic action scene in a crowded traditional Asian wet market during heavy rain and severe flooding. A young Asian woman with long wet dark hair, wearing a pink waterproof rain suit and black rubber boots, runs through flooded market aisles while a massive wave of water rushes behind her. Show seafood and vegetable stalls, floating crates, overturned objects, panicked crowds, rushing dirty water, reflections, splashes and realistic wet details. Use fast handheld tracking, close-ups, low-angle water shots, side tracking and wide shots, with realistic physics, motion blur and dynamic camera shake. She jumps over obstacles and runs across raised market surfaces. Near the end, she reaches a raised area where a frightened cow is trapped by the flood. She looks at it and urgently says: “Come on, girl! Move!” She guides the cow toward safety. Keep her face, hairstyle and pink outfit consistent. Dark rainy atmosphere, cool blue-gray tones mixed with warm fluorescent lights, cinematic depth of field, volumetric lighting, realistic skin, water and reflections, subtle film grain, 4K anamorphic Hollywood action style. Add large white brush-stroke typography “RUN” briefly across the center. No cartoon look, CGI appearance, distorted anatomy, extra fingers, duplicated people or face changes.

Noor

17,446 次观看 • 20 天前

The UFOs were here again tonight! I was out there with them and just came back in. Something is going on! They’re going to do another big event soon, or something big is about to happen in the world! This isn’t normal—they normally don't come this many times so close together. There were so many tonight I couldn’t count them all! I’ll show everyone what they look like at a distance with the plasma field off. I have them on other cameras doing this too, but the color Starlight camera shows it best. They blink really fast when in large groups like tonight. I don't know if they do that to try and mimic aircraft or why they do it. I see them very closeup here and they are nothing like an aircraft! I use radar apps and satellite tracking apps so I know what everything is flying all around this area. I have been doing this since 2005. I also track the military aircraft that don't show up on radar and I have cameras that can zoom in on them miles away. The UFOs are blinking way faster than any aircraft navigation lights blink when the plasma field is not glowing. They don't blink like this all the time when the plasma is not glowing. They do it many times when in the larger groups for some reason. Some also turned their plasma fields on. Two C-17s and a C-130 flew over on a training mission shortly before the UFOs appeared and were still in the area, so there’s a good chance some USAF pilots saw them tonight. Some of the SOLL pilots have seen them here before over the years when the UFOs jump in behind or above them or near them. They never harm the aircraft. We’ve seen them approach SOLL aircraft many times over the years when the aircraft fly over at tree top level training here and they cause no harm to them. I have the UFOs with the SOLL aircraft on videos together. They just let them know they are here. One of the sphere UFO appeared right in front of a Life Star helicopter beside our house also; the pilot saw it, pulled back, and the UFO instantly jumped away without touching the helicopter or harming it. The small lights higher up crossing the sky are Starlink satellites and there are a few aircraft crossing higher up. The UFOs are all the reddish orange lights blinking fast near the tree line and you can see some of them light the plasma field up. One of the UFOs fly's in a circle pattern while the plasma is lit up.

The UFOs were here again tonight! I was out there with them and just came back in. Something is going on! They’re going to do another big event soon, or something big is about to happen in the world! This isn’t normal—they normally don't come this many times so close together. There were so many tonight I couldn’t count them all! I’ll show everyone what they look like at a distance with the plasma field off. I have them on other cameras doing this too, but the color Starlight camera shows it best. They blink really fast when in large groups like tonight. I don't know if they do that to try and mimic aircraft or why they do it. I see them very closeup here and they are nothing like an aircraft! I use radar apps and satellite tracking apps so I know what everything is flying all around this area. I have been doing this since 2005. I also track the military aircraft that don't show up on radar and I have cameras that can zoom in on them miles away. The UFOs are blinking way faster than any aircraft navigation lights blink when the plasma field is not glowing. They don't blink like this all the time when the plasma is not glowing. They do it many times when in the larger groups for some reason. Some also turned their plasma fields on. Two C-17s and a C-130 flew over on a training mission shortly before the UFOs appeared and were still in the area, so there’s a good chance some USAF pilots saw them tonight. Some of the SOLL pilots have seen them here before over the years when the UFOs jump in behind or above them or near them. They never harm the aircraft. We’ve seen them approach SOLL aircraft many times over the years when the aircraft fly over at tree top level training here and they cause no harm to them. I have the UFOs with the SOLL aircraft on videos together. They just let them know they are here. One of the sphere UFO appeared right in front of a Life Star helicopter beside our house also; the pilot saw it, pulled back, and the UFO instantly jumped away without touching the helicopter or harming it. The small lights higher up crossing the sky are Starlink satellites and there are a few aircraft crossing higher up. The UFOs are all the reddish orange lights blinking fast near the tree line and you can see some of them light the plasma field up. One of the UFOs fly's in a circle pattern while the plasma is lit up.

Ranger H

17,563 次观看 • 9 个月前

Created this race using GPT Image 2 and Seedance 2.0 on TapNow Prompt Follow the storyboard strictly in exact order from Panel 1 to Panel 9. Do not skip, merge, or rearrange scenes. Keep the SAME female cyclist identity across the entire film. No face changes, no hairstyle changes, no helmet changes, no body proportion inconsistencies. Baby pink must remain the dominant apparel color throughout all cycling scenes. Avoid black wardrobe replacements. Preserve realistic nighttime lighting continuity between shots. Maintain the same cool blue tones and subtle red light reflections. Heavy rain intensity must stay visually consistent across all scenes. Water physics must look physically accurate: droplets, splashes, mist, wheel spray, and runoff should behave naturally. Avoid artificial AI motion. Camera movement should feel like real cinema rigs, FPV drones, mounted bike cameras, or stabilized tracking systems. Drone shots must maintain locked framing and smooth movement without random drifting or orbiting. Use subtle cinematic motion only — no excessive shaking or jitter. Keep realistic breathing, body fatigue, pedaling mechanics, and fabric reactions to wind and rain. Preserve shallow depth of field in macro shots and atmospheric haze in wide shots. Keep the environment dark, moody, and cinematic with strong contrast between wet reflections and darkness. Ensure all reflections on asphalt, water droplets, and bike components react naturally to changing light sources. Maintain premium commercial pacing: slow controlled preparation and macro shots transitioning into aggressive high-speed riding sequences. Final output should resemble a high-budget Nike / Rapha night cycling commercial shot during a real mountain storm. Ultra-realistic cinematic night cycling commercial about female endurance cyclists riding through an intense rainstorm in the mountains at night. Premium Nike / Rapha aesthetic with baby pink performance cycling apparel as the dominant accent color. Hyper-realistic documentary look, no stylization, no anime look, no beauty filters. Natural skin texture, realistic rain interaction, physically accurate water behavior, cinematic low-key lighting, cool blue night tones mixed with subtle red rear-light reflections. Heavy rain, fog, wet asphalt reflections, cinematic motion blur, high dynamic range, shallow depth of field, premium sports commercial quality. The film follows a strict 9-panel storyboard structure with seamless cinematic transitions and continuity preserved across every scene. The SAME female cyclist identity must remain consistent throughout the entire video: same face, helmet, glasses, body proportions, baby pink apparel, lighting style, and overall appearance. Maintain continuity of rain intensity, wetness, fog density, and environmental lighting between all shots. Panel 1: Extreme macro close-up of the female cyclist’s eyes and face in heavy rain at night. Focus on soaked eyelashes, wet skin texture, raindrops streaming across the face, baby pink helmet and baby pink face mask visible. Red rear bike light flickers dynamically across her eyes and skin while cool blue night tones dominate the scene. High contrast cinematic lighting, shallow depth of field, subtle breathing motion, intense determined expression. Panel 2: Cinematic medium close-up frontal shot of the cyclist riding aggressively through heavy rain at night. She pedals hard with strong effort and forward-leaning posture. Baby pink waterproof cycling jacket soaked with rainwater. Front bike light cuts through fog and rain with subtle flickering illumination. Wet asphalt reflects red and white lights. Smooth cinematic tracking shot with controlled stable motion and slight natural float. Panel 3: Ultra-realistic macro shot of large raindrops impacting wet asphalt at night. Crown-shaped splashes and overlapping ripples in slow motion. Rough wet asphalt texture, cool blue cinematic tones, subtle reflections from bike lights.

Created this race using GPT Image 2 and Seedance 2.0 on TapNow Prompt Follow the storyboard strictly in exact order from Panel 1 to Panel 9. Do not skip, merge, or rearrange scenes. Keep the SAME female cyclist identity across the entire film. No face changes, no hairstyle changes, no helmet changes, no body proportion inconsistencies. Baby pink must remain the dominant apparel color throughout all cycling scenes. Avoid black wardrobe replacements. Preserve realistic nighttime lighting continuity between shots. Maintain the same cool blue tones and subtle red light reflections. Heavy rain intensity must stay visually consistent across all scenes. Water physics must look physically accurate: droplets, splashes, mist, wheel spray, and runoff should behave naturally. Avoid artificial AI motion. Camera movement should feel like real cinema rigs, FPV drones, mounted bike cameras, or stabilized tracking systems. Drone shots must maintain locked framing and smooth movement without random drifting or orbiting. Use subtle cinematic motion only — no excessive shaking or jitter. Keep realistic breathing, body fatigue, pedaling mechanics, and fabric reactions to wind and rain. Preserve shallow depth of field in macro shots and atmospheric haze in wide shots. Keep the environment dark, moody, and cinematic with strong contrast between wet reflections and darkness. Ensure all reflections on asphalt, water droplets, and bike components react naturally to changing light sources. Maintain premium commercial pacing: slow controlled preparation and macro shots transitioning into aggressive high-speed riding sequences. Final output should resemble a high-budget Nike / Rapha night cycling commercial shot during a real mountain storm. Ultra-realistic cinematic night cycling commercial about female endurance cyclists riding through an intense rainstorm in the mountains at night. Premium Nike / Rapha aesthetic with baby pink performance cycling apparel as the dominant accent color. Hyper-realistic documentary look, no stylization, no anime look, no beauty filters. Natural skin texture, realistic rain interaction, physically accurate water behavior, cinematic low-key lighting, cool blue night tones mixed with subtle red rear-light reflections. Heavy rain, fog, wet asphalt reflections, cinematic motion blur, high dynamic range, shallow depth of field, premium sports commercial quality. The film follows a strict 9-panel storyboard structure with seamless cinematic transitions and continuity preserved across every scene. The SAME female cyclist identity must remain consistent throughout the entire video: same face, helmet, glasses, body proportions, baby pink apparel, lighting style, and overall appearance. Maintain continuity of rain intensity, wetness, fog density, and environmental lighting between all shots. Panel 1: Extreme macro close-up of the female cyclist’s eyes and face in heavy rain at night. Focus on soaked eyelashes, wet skin texture, raindrops streaming across the face, baby pink helmet and baby pink face mask visible. Red rear bike light flickers dynamically across her eyes and skin while cool blue night tones dominate the scene. High contrast cinematic lighting, shallow depth of field, subtle breathing motion, intense determined expression. Panel 2: Cinematic medium close-up frontal shot of the cyclist riding aggressively through heavy rain at night. She pedals hard with strong effort and forward-leaning posture. Baby pink waterproof cycling jacket soaked with rainwater. Front bike light cuts through fog and rain with subtle flickering illumination. Wet asphalt reflects red and white lights. Smooth cinematic tracking shot with controlled stable motion and slight natural float. Panel 3: Ultra-realistic macro shot of large raindrops impacting wet asphalt at night. Crown-shaped splashes and overlapping ripples in slow motion. Rough wet asphalt texture, cool blue cinematic tones, subtle reflections from bike lights.

Sharon Riley

72,716 次观看 • 4 个月前

Here is some videos of the larger craft from 12-4-2024 and some of the UFO Spheres from 12-3-2024. The Spheres have been here every night this week. They have been down low by the house like normal but we have seen many of them flying all around the area very fast. They will blink very fast or keep the plasma field on but at a lower brightness than when they are just hovering beside the house. On 12-4-2024 we were watching the Spheres flying all around then we saw the larger UFO appear after a helicopter circled the area. It stayed beside the house for over an hour only moving about 800 meters in that amount of time. It was under 1000 feet in altitude and bigger than C-17 aircraft. It looked saucer shaped but wider in the center and tapered around the edges. It was showing up very hot on the thermal camera even though no visible light at all was coming from it. The Spheres showed up cold on the thermal camera when the plasma field was not glowing when we filmed them closer up before. This larger craft had the plasma field around it just like the Spheres but it was not glowing. We could not see the craft with our naked eyes or on the starlight camera or IR cameras. Only the thermal camera would pick it up. Can see how the air around it is shimmering and can see the field around it. It looked better in person but what I filmed is important because you all will see this again on some daytime close up videos! I have watched these craft up close many times and they have this plasma field around them and even when its not glowing its there. In the daytime you can see the air shimmering around the craft very good. We have seen them use this to bend the light around the craft someway and turn the craft almost completely invisible but you could still see the air shimmering around the craft. I had some cheap starlight cameras that would pick the larger crafts up when the Spheres were beside them glowing and the larger craft was almost invisible. The glow from the Spheres plasma would outline were the larger craft was and we could see it on those starlight cameras. We had a handheld digital starlight camera when this craft was over us and it would not pick it up at all. This is the real UFOs and we have seen them here for over twenty years and after the close encounters in 2012 they started staying very close to us and showing up very often. Even when we cant see them we have seen things to let us know they are near us. The Spheres can change shape and we have them on videos doing it. The larger craft we have seen have all looked different but the Spheres are a constant even though they can change shape. During the close encounter on 11-8-2012 there were many of the Spheres and two larger U shaped craft that the Beings came out of. On 2-16-2023 there was a larger craft here that was over 300 meters long and is the biggest craft we have seen the Spheres with. Then we saw this craft that we have seen before a few times including in the daytime. The Spheres have been with each of the larger craft every time. The Spheres act like they guard and protect the larger craft each time they are around. Sometimes they stay very close to them and on 2-16-2023 some of the Spheres were going in the bottom of the giant craft or attaching to the bottom of it and there was an electrical charge or something that would go all around the larger craft when the Spheres attached or went into it. When this larger craft on the 4th was sitting there we saw several of the Spheres flying around really fast with the plasma field on a constant low glow. We have seen them doing this all around the area this whole week. The other ones were down low beside the house and that's were they normally stay close to when they show themselves. I'm not sure what the ones flying all around were doing. Its really different activity the way they were acting. Monday there were so many doing this in every direction we could not count them all. It had to be in the hundreds. We watch the radars also and watch all the aircraft around us during these events and none of the craft we were seeing shows on any public radar sites. The military radar site does not show them either. Monday there was a large increase in commercial fights and some of these aircraft had to see all the smaller craft flying around really fast. Some had the plasma glowing constant but many had it flashing really fast. Way faster than the navigation lights of regular aircraft. Many of the so called drones being seen around the world and in America the last few weeks are these craft and there are some videos going around from New Jersey that shows these craft but some are showing something that looks more man made. On the ones that look more man made there has been sound that sounds like jet engines on many I have seen. The craft here make ZERO sound most of the time! We have only heard a few very strange sounds come from them over the years. One sound sounded like a giant transformer humming during the close encounter on 11-8-2012 and on 2-16-2023 we heard what sounded like very loud trumpets come from the sky! We were not the only ones that heard that. People up the road that can not see this area and that knows nothing about what's going on here told us they heard the trumpet sounds and thought it was the end of the world and told us they started praying! They also told us they saw a light floating through the trees back toward our land. There has been some other strange sounds also and something really strange is in the summer time there are hundreds of thousand of frogs in the swamps around our land when its raining more and when the UFOs show up all the frogs will stop singing at one time. This is on some of the videos I have. When you are out there beside the swamps and there are that many frogs singing then they all stop at one time its very strange! Its not just the frogs, all the animals and insects seem like they stop. Sometimes they do it right before the plasma field turns on and sometimes right after. Two of my male dogs would go crazy also right before the UFOs show themselves. I have other dogs but those two were super sensitive to whatever was coming from the UFOs. I think they all are hearing something from the UFOs that we cant hear or they are feeling what I feel from them when they are closer to me. Either way its probably something that scientist could pick up with the right equipment. #UFO #UAP #RealUFOs #Spheres #ufox #ua

Here is some videos of the larger craft from 12-4-2024 and some of the UFO Spheres from 12-3-2024. The Spheres have been here every night this week. They have been down low by the house like normal but we have seen many of them flying all around the area very fast. They will blink very fast or keep the plasma field on but at a lower brightness than when they are just hovering beside the house. On 12-4-2024 we were watching the Spheres flying all around then we saw the larger UFO appear after a helicopter circled the area. It stayed beside the house for over an hour only moving about 800 meters in that amount of time. It was under 1000 feet in altitude and bigger than C-17 aircraft. It looked saucer shaped but wider in the center and tapered around the edges. It was showing up very hot on the thermal camera even though no visible light at all was coming from it. The Spheres showed up cold on the thermal camera when the plasma field was not glowing when we filmed them closer up before. This larger craft had the plasma field around it just like the Spheres but it was not glowing. We could not see the craft with our naked eyes or on the starlight camera or IR cameras. Only the thermal camera would pick it up. Can see how the air around it is shimmering and can see the field around it. It looked better in person but what I filmed is important because you all will see this again on some daytime close up videos! I have watched these craft up close many times and they have this plasma field around them and even when its not glowing its there. In the daytime you can see the air shimmering around the craft very good. We have seen them use this to bend the light around the craft someway and turn the craft almost completely invisible but you could still see the air shimmering around the craft. I had some cheap starlight cameras that would pick the larger crafts up when the Spheres were beside them glowing and the larger craft was almost invisible. The glow from the Spheres plasma would outline were the larger craft was and we could see it on those starlight cameras. We had a handheld digital starlight camera when this craft was over us and it would not pick it up at all. This is the real UFOs and we have seen them here for over twenty years and after the close encounters in 2012 they started staying very close to us and showing up very often. Even when we cant see them we have seen things to let us know they are near us. The Spheres can change shape and we have them on videos doing it. The larger craft we have seen have all looked different but the Spheres are a constant even though they can change shape. During the close encounter on 11-8-2012 there were many of the Spheres and two larger U shaped craft that the Beings came out of. On 2-16-2023 there was a larger craft here that was over 300 meters long and is the biggest craft we have seen the Spheres with. Then we saw this craft that we have seen before a few times including in the daytime. The Spheres have been with each of the larger craft every time. The Spheres act like they guard and protect the larger craft each time they are around. Sometimes they stay very close to them and on 2-16-2023 some of the Spheres were going in the bottom of the giant craft or attaching to the bottom of it and there was an electrical charge or something that would go all around the larger craft when the Spheres attached or went into it. When this larger craft on the 4th was sitting there we saw several of the Spheres flying around really fast with the plasma field on a constant low glow. We have seen them doing this all around the area this whole week. The other ones were down low beside the house and that's were they normally stay close to when they show themselves. I'm not sure what the ones flying all around were doing. Its really different activity the way they were acting. Monday there were so many doing this in every direction we could not count them all. It had to be in the hundreds. We watch the radars also and watch all the aircraft around us during these events and none of the craft we were seeing shows on any public radar sites. The military radar site does not show them either. Monday there was a large increase in commercial fights and some of these aircraft had to see all the smaller craft flying around really fast. Some had the plasma glowing constant but many had it flashing really fast. Way faster than the navigation lights of regular aircraft. Many of the so called drones being seen around the world and in America the last few weeks are these craft and there are some videos going around from New Jersey that shows these craft but some are showing something that looks more man made. On the ones that look more man made there has been sound that sounds like jet engines on many I have seen. The craft here make ZERO sound most of the time! We have only heard a few very strange sounds come from them over the years. One sound sounded like a giant transformer humming during the close encounter on 11-8-2012 and on 2-16-2023 we heard what sounded like very loud trumpets come from the sky! We were not the only ones that heard that. People up the road that can not see this area and that knows nothing about what's going on here told us they heard the trumpet sounds and thought it was the end of the world and told us they started praying! They also told us they saw a light floating through the trees back toward our land. There has been some other strange sounds also and something really strange is in the summer time there are hundreds of thousand of frogs in the swamps around our land when its raining more and when the UFOs show up all the frogs will stop singing at one time. This is on some of the videos I have. When you are out there beside the swamps and there are that many frogs singing then they all stop at one time its very strange! Its not just the frogs, all the animals and insects seem like they stop. Sometimes they do it right before the plasma field turns on and sometimes right after. Two of my male dogs would go crazy also right before the UFOs show themselves. I have other dogs but those two were super sensitive to whatever was coming from the UFOs. I think they all are hearing something from the UFOs that we cant hear or they are feeling what I feel from them when they are closer to me. Either way its probably something that scientist could pick up with the right equipment. #UFO #UAP #RealUFOs #Spheres #ufox #ua

Ranger H

39,660 次观看 • 1 年前

Prompt: Create a 43-second cinematic, photorealistic action-comedy sequence inside an old, cramped, slightly abandoned apartment/bathroom with realistic worn walls, tiled surfaces, warm practical lighting, and natural shadows. A young adult man in a light gray T-shirt is standing in front of a bathroom mirror. Suddenly, another man wearing a black-and-white striped shirt appears and aggressively grabs him. The two struggle physically through the small bathroom, with realistic body movement, pushing, pulling, stumbling and fighting. The action continues dynamically through the bathroom. One character crashes against the wall and breaks through a damaged section, creating a large opening into the adjacent room. Dust and small debris fall naturally as the wall breaks. Continue with a fast-paced chase through the apartment. The gray-shirted man runs toward a window while the striped-shirted man follows. Use handheld camera movement, realistic motion blur, quick tracking shots and dramatic close-ups to make the sequence feel like a real action movie. The characters then escape outside into an old brick courtyard/alley. One character climbs onto a rusty metal fire escape while the other follows from below. Show realistic physics, footsteps, clothing movement and environmental interaction. Final scene: transition to a large outdoor dumpster filled with cardboard and garbage. The gray-shirted man is inside the dumpster, looking exhausted and confused, holding a small green plush toy. He slowly looks around as the camera pushes in toward his face. End on a slightly comedic reaction shot. Visual style: ultra-photorealistic, cinematic action film, realistic human anatomy, natural skin texture, detailed environments, practical lighting, subtle film grain, realistic physics, believable facial expressions. Camera: dynamic handheld cinematography, wide establishing shots, medium tracking shots, close-ups during the struggle, smooth camera transitions, natural depth of field, 35mm/50mm cinematic lenses. Motion: realistic human movement, accurate weight and momentum, natural collisions, realistic debris and dust simulation, no slow-motion unless used very briefly for dramatic impact. Important: Maintain consistent character appearance, clothing, hairstyle and body proportions throughout the entire video. No random character changes, no extra limbs, no distorted faces, no cartoon appearance, no text, no subtitles, no watermark.

Prompt: Create a 43-second cinematic, photorealistic action-comedy sequence inside an old, cramped, slightly abandoned apartment/bathroom with realistic worn walls, tiled surfaces, warm practical lighting, and natural shadows. A young adult man in a light gray T-shirt is standing in front of a bathroom mirror. Suddenly, another man wearing a black-and-white striped shirt appears and aggressively grabs him. The two struggle physically through the small bathroom, with realistic body movement, pushing, pulling, stumbling and fighting. The action continues dynamically through the bathroom. One character crashes against the wall and breaks through a damaged section, creating a large opening into the adjacent room. Dust and small debris fall naturally as the wall breaks. Continue with a fast-paced chase through the apartment. The gray-shirted man runs toward a window while the striped-shirted man follows. Use handheld camera movement, realistic motion blur, quick tracking shots and dramatic close-ups to make the sequence feel like a real action movie. The characters then escape outside into an old brick courtyard/alley. One character climbs onto a rusty metal fire escape while the other follows from below. Show realistic physics, footsteps, clothing movement and environmental interaction. Final scene: transition to a large outdoor dumpster filled with cardboard and garbage. The gray-shirted man is inside the dumpster, looking exhausted and confused, holding a small green plush toy. He slowly looks around as the camera pushes in toward his face. End on a slightly comedic reaction shot. Visual style: ultra-photorealistic, cinematic action film, realistic human anatomy, natural skin texture, detailed environments, practical lighting, subtle film grain, realistic physics, believable facial expressions. Camera: dynamic handheld cinematography, wide establishing shots, medium tracking shots, close-ups during the struggle, smooth camera transitions, natural depth of field, 35mm/50mm cinematic lenses. Motion: realistic human movement, accurate weight and momentum, natural collisions, realistic debris and dust simulation, no slow-motion unless used very briefly for dramatic impact. Important: Maintain consistent character appearance, clothing, hairstyle and body proportions throughout the entire video. No random character changes, no extra limbs, no distorted faces, no cartoon appearance, no text, no subtitles, no watermark.

Karlos

23,997 次观看 • 18 天前

ARE WE ALONE? Harvard’s Avi Loeb is running the UAP Science Advisory Council, and he’s not here for the hype. Lara Trump sat down with him at Harvard to separate fact from fiction. Five major tranches of Pentagon data have now been released. Objects are still showing up over military installations and strategic sites. Loeb’s take is straightforward: “If we see objects hovering over strategic assets and we don’t know what they are, we need to figure them out. That’s common sense.” He reviewed the newly declassified videos with us. Some show objects moving in ways that don’t match known drones or U.S. missiles. In one case where they could calculate speed using background terrain, it tracked like a missile, but no U.S. system matched the shape or performance exactly. The council’s recommendation? Deploy better sensors. Multiple cameras for distance and tracking. Spectroscopy to analyze what these orbs are made of. Instruments that can tell us composition if any material is recovered. Loeb’s team requested more than 50 specific data items. Some remain classified because of the sensors used — not necessarily the objects themselves. On the big question — are these alien? “We haven’t seen clear evidence for alien technology yet.” But he refuses to dismiss the possibility. The sun is relatively young. Most stars formed long before ours. It’s completely reasonable, he says, to assume more advanced civilizations exist. Loeb co-founded the Galileo Project, which is actively scanning the skies with telescopes (including one near the Sphere in Las Vegas). So far, nothing unexplained. His standing order to the team: “If you see anything, wake me up in the middle of the night.” His bottom line cuts through both the skeptics and the true believers: “The foundation of science is curiosity. Not claiming we already know the answer.” What are we actually seeing in our skies? The Trump administration opened the door. Loeb’s council is trying to answer it with data instead of dogma. The search is on.

ARE WE ALONE? Harvard’s Avi Loeb is running the UAP Science Advisory Council, and he’s not here for the hype. Lara Trump sat down with him at Harvard to separate fact from fiction. Five major tranches of Pentagon data have now been released. Objects are still showing up over military installations and strategic sites. Loeb’s take is straightforward: “If we see objects hovering over strategic assets and we don’t know what they are, we need to figure them out. That’s common sense.” He reviewed the newly declassified videos with us. Some show objects moving in ways that don’t match known drones or U.S. missiles. In one case where they could calculate speed using background terrain, it tracked like a missile, but no U.S. system matched the shape or performance exactly. The council’s recommendation? Deploy better sensors. Multiple cameras for distance and tracking. Spectroscopy to analyze what these orbs are made of. Instruments that can tell us composition if any material is recovered. Loeb’s team requested more than 50 specific data items. Some remain classified because of the sensors used — not necessarily the objects themselves. On the big question — are these alien? “We haven’t seen clear evidence for alien technology yet.” But he refuses to dismiss the possibility. The sun is relatively young. Most stars formed long before ours. It’s completely reasonable, he says, to assume more advanced civilizations exist. Loeb co-founded the Galileo Project, which is actively scanning the skies with telescopes (including one near the Sphere in Las Vegas). So far, nothing unexplained. His standing order to the team: “If you see anything, wake me up in the middle of the night.” His bottom line cuts through both the skeptics and the true believers: “The foundation of science is curiosity. Not claiming we already know the answer.” What are we actually seeing in our skies? The Trump administration opened the door. Loeb’s council is trying to answer it with data instead of dogma. The search is on.

Gunther Eagleman™

47,851 次观看 • 28 天前

$After 8 months of building in stealth and testing our infrastructure on 10000+ hours of real-world data and hundreds of unique environments, we're bringing FPV Labs into the open today. FPV Labs started with the following bet - if human data proves to be the underlying factor that determines scaling laws in general-purpose robotics, it will trigger the largest economic transformation in human history, and the underlying infrastructure that captures that data will determine how fast we get there. We will achieve this by building the full-stack infrastructure for capturing, processing, transferring, and evaluating human experience into spatial, temporal, and semantic knowledge for machines. Despite all the research novelty behind ChatGPT, its success can be attributed to one foundational fact - the scaling law of transformers. We believe the same dynamics have made their way into robotics. Recent studies showed task completion rates jumping from 30% to 70% when human demonstration data scaled from 1,000 to 20,000 hours, a log-linear trend that mirrors exactly what we saw in language and vision. Seeing these emergent signs of scaling law curves in robotics, we believe we are entering the era of general-purpose robotics policies, which makes the next few years the most exciting time in the history of this field. But the library of physical interactions required to train general-purpose robot policies does not exist yet. Over the last 8 months, we've seen dozens of companies emerge in this space. We were really happy to see new companies pushing this space forward, but we also saw the same pattern repeat: every egocentric data company was making some tradeoffs between quality, scale, and diversity. We have built FPV labs on the core principle that high-quality data is orders of magnitude more valuable than sheer volume. Case in point, self-driving cars collect thousands of hours of data per day, but only a small fraction of that data is actually useful for training better models. Several studies, like RT-2, have shown that as little as 1% of data improves as much as 25% on task success. The quality and diversity of data matter a lot more than scale, so there is clearly a power law curve in the downstream impact of data. We've spent months obsessing over data quality by building our stack, discarding it, rebuilding it, and iterating until we found a formula that doesn't compromise downstream quality at scale. We believe the downstream impact here is far more profound than most people realize. Workers globally are paid around $60 trillion per year in aggregate, and a lion's share of that compensation goes to physical labor - tasks that require navigating real spaces, manipulating real objects, and negotiating the infinite variability of the physical world. Human-to-robot transfer will be one of the most important infrastructures that will shape our society in the near future, and if it works, the economic impact will dwarf every technology transition that came before it in an exponential manner and lead to the creation of goods and services we can’t imagine today. Our mission is to lay the groundwork for us to transition into this future - the future of abundance. We are deeply grateful to our earliest believers, Paras Chopra and Lossfunk, who played a critical role in shaping our thinking.$

After 8 months of building in stealth and testing our infrastructure on 10000+ hours of real-world data and hundreds of unique environments, we're bringing FPV Labs into the open today. FPV Labs started with the following bet - if human data proves to be the underlying factor that determines scaling laws in general-purpose robotics, it will trigger the largest economic transformation in human history, and the underlying infrastructure that captures that data will determine how fast we get there. We will achieve this by building the full-stack infrastructure for capturing, processing, transferring, and evaluating human experience into spatial, temporal, and semantic knowledge for machines. Despite all the research novelty behind ChatGPT, its success can be attributed to one foundational fact - the scaling law of transformers. We believe the same dynamics have made their way into robotics. Recent studies showed task completion rates jumping from 30% to 70% when human demonstration data scaled from 1,000 to 20,000 hours, a log-linear trend that mirrors exactly what we saw in language and vision. Seeing these emergent signs of scaling law curves in robotics, we believe we are entering the era of general-purpose robotics policies, which makes the next few years the most exciting time in the history of this field. But the library of physical interactions required to train general-purpose robot policies does not exist yet. Over the last 8 months, we've seen dozens of companies emerge in this space. We were really happy to see new companies pushing this space forward, but we also saw the same pattern repeat: every egocentric data company was making some tradeoffs between quality, scale, and diversity. We have built FPV labs on the core principle that high-quality data is orders of magnitude more valuable than sheer volume. Case in point, self-driving cars collect thousands of hours of data per day, but only a small fraction of that data is actually useful for training better models. Several studies, like RT-2, have shown that as little as 1% of data improves as much as 25% on task success. The quality and diversity of data matter a lot more than scale, so there is clearly a power law curve in the downstream impact of data. We've spent months obsessing over data quality by building our stack, discarding it, rebuilding it, and iterating until we found a formula that doesn't compromise downstream quality at scale. We believe the downstream impact here is far more profound than most people realize. Workers globally are paid around $60 trillion per year in aggregate, and a lion's share of that compensation goes to physical labor - tasks that require navigating real spaces, manipulating real objects, and negotiating the infinite variability of the physical world. Human-to-robot transfer will be one of the most important infrastructures that will shape our society in the near future, and if it works, the economic impact will dwarf every technology transition that came before it in an exponential manner and lead to the creation of goods and services we can’t imagine today. Our mission is to lay the groundwork for us to transition into this future - the future of abundance. We are deeply grateful to our earliest believers, Paras Chopra and Lossfunk, who played a critical role in shaping our thinking.

Abhishek Anand

82,678 次观看 • 5 个月前

🚀Announcing NeRSemble 3D Head Avatar Benchmark v2 Version 2 of the NeRSemble 3D Head Avatar Benchmark systematically evaluates several aspects of 3D head avatar creation. Our goal is to drive progress toward more realistic, robust, and generalizable avatar methods. 🔬Benchmark Tasks The NeRSemble Benchmark v2 features three core challenges: - Dynamic Novel View Synthesis - Monocular FLAME-driven Avatar Creation (updated) - Single-view 3D Face Reconstruction (new) 👉Explore the online leaderboard and submission system: 🆕What's new? 1. New Task: Single-view 3D Face Reconstruction Given a single portrait image, reconstruct an accurate 3D mesh either showing the input expression or a fully neutral one. Unlike prior benchmarks, the NeRSemble benchmark emphasizes diverse and challenging facial expressions, better reflecting real scenarios. For technical details, see the Pixel3DMM paper. 2. Updated task: Monocular FLAME-driven Avatar Creation We have improved the FLAME tracking that is used for both avatar creation from the monocular videos and avatar driving on the hidden test sequences. The updated benchmark task has: - more stable torso tracking - more expressive lip closures during speech - Improved mouth tracking for challenging facial expressions We hope that these improvements to the benchmark help drive the field forward. 🏆 CVPR 2026 Workshop & Prizes The NeRSemble benchmark will be featured at the CVPR 2026 Workshop on Photo-realistic 3D Head Avatars. Participants in the new and updated tasks have the opportunity to win: - 🎁RTX 5080 GPUs (sponsored by NVIDIA) - 🎤15-minute oral presentation at the workshop ⏰ Submission Deadline - May 26, 2026 Reach out to the amazing Tobias Kirschstein and Simon Giebenhain for more details :)

Matthias Niessner

30,086 次观看 • 5 个月前

She was just the "damsel in distress" Night training turned into a full-blown awakening — one wrong move and this girl turned campus grounds into a battlefield. Fists, flames, and a final blow you won't see coming. Made with Seedance 2.5 on Fish Creative Prompt: Core Theme University field, one take fight scene. Two fighters, hand to hand combat with real stunts and kicks. Fight ends with the girl's fire power awakening. She finishes him with a decisive final blow. Continuous shot, no cuts, no edits. 16:9 aspect ratio. Character Appearance Lock FIGHTER A (Boy) Asian male, around 20 years old. Short black hair, slightly covering forehead. Almond-shaped brown eyes, defined jawline. Fitted white athletic tank top. Yellow lemon-colored shorts, black sneakers. Confident, relaxed posture, athletic build. FIGHTER B (Girl) Japanese female, early 20s, striking model-like features. Lean, tall, athletic figure. Silver-lavender hair, chin length bob with straight fringe. Sharp, focused amber eyes. Pink acrylic short skirt, high-shine finish. Cropped white top. Pink knee-high socks, white sneakers. Fast, agile, aggressive fighting style — real kicks, spins, and strikes. Environment Lock University open ground, trimmed grass field with dirt patches. White painted field lines. Overhead campus lamp posts casting pools of light. Everything outside the light pools is deep shadow. University buildings faintly visible in the background darkness. Dust and grass kick up on every impact, roll, and landing. Timeline 0.0–2.5s: Fighter A stands behind Fighter B. Right palm covers her mouth from behind. She grabs at his wrist, eyes sharp and furious. She drives her elbow into his ribs, breaks free, spins out. She lands in a fighting stance, dust rising. 2.5–5.0s: She rolls her shoulders, cracks her neck. Walks forward with slow, aggressive confidence. Speaks clearly in English to camera: "You really thought you could catch me off guard?" 5.0–7.0s: He sprints forward, throws a right hook. She ducks low, sweeps his front leg. He stumbles, catches balance, resets stance. 7.0–9.0s: He throws a straight right jab. She parries with her left forearm, steps in. Delivers a sharp right knee strike to his side. He staggers back two steps. 9.0–12.5s: He lands a glancing punch on her shoulder. She barely reacts, grabs his wrist, twists. Executes a real hip-throw, slams him to the ground. He rolls once, pushes up onto one knee, breathing hard. 12.5–16.0s: Camera pushes in on her face. Sweat, dirt, sharp determined eyes. First orange-red flame threads flicker along her fingertips. Threads crawl up her arms to her shoulders. Her eyes begin to glow amber-orange from within. The lamp posts start flickering irregularly. 16.0–18.0s: She exhales sharply, rolls her wrists. Small flames bloom across both palms. Fabric of her top singes faintly at the edges. Heat distortion ripples the air around her. 18.0–21.0s: Camera circles her fast. He rises, throws a desperate punch. She sidesteps, grabs his arm, flame licking her grip. Grass scorches in a ring beneath her feet. Flame trails follow every movement of her limbs. 21.0–23.0s: She unleashes a flurry — spinning back kick, followed by a jumping roundhouse. Each strike leaves a brief trail of fire in the air. He blocks two, takes the third across the jaw. He drops to one knee, dazed, smoke rising off his shoulder. 23.0–27.5s: She plants her feet, pulls both fists back, flames coiling tight around her forearms. Steps in with full aggression, drives a flaming knee strike upward into his chest. Impact releases a burst of fire and sparks outward. 27.5–30.0s: He's launched back, lands hard, skidding through dirt. She stands over him, flames dying down her arms to embers. Breathing hard, one fist still smoking. Lamp posts dim to near darkness. Only the fading orange glow lights her face. Then everything goes dark. Lighting Extremely dark night, deep shadows. Lamp posts as rim light on shoulders and hair. Warm orange-red glow after her power appears, casting light on both fighters. Sound Design Location sound only. Grass and dirt scraping, fabric rustling. Sharp impact sounds, heavy breathing, grunts. Flame ignition crackle building to a low roar. Lamp buzz as they flicker. Her line delivered clearly at 2.5–5.0s. Sharp exhale and crackle at the final blow. Total silence at the very end. No background music throughout.

She was just the "damsel in distress" Night training turned into a full-blown awakening — one wrong move and this girl turned campus grounds into a battlefield. Fists, flames, and a final blow you won't see coming. Made with Seedance 2.5 on Fish Creative Prompt: Core Theme University field, one take fight scene. Two fighters, hand to hand combat with real stunts and kicks. Fight ends with the girl's fire power awakening. She finishes him with a decisive final blow. Continuous shot, no cuts, no edits. 16:9 aspect ratio. Character Appearance Lock FIGHTER A (Boy) Asian male, around 20 years old. Short black hair, slightly covering forehead. Almond-shaped brown eyes, defined jawline. Fitted white athletic tank top. Yellow lemon-colored shorts, black sneakers. Confident, relaxed posture, athletic build. FIGHTER B (Girl) Japanese female, early 20s, striking model-like features. Lean, tall, athletic figure. Silver-lavender hair, chin length bob with straight fringe. Sharp, focused amber eyes. Pink acrylic short skirt, high-shine finish. Cropped white top. Pink knee-high socks, white sneakers. Fast, agile, aggressive fighting style — real kicks, spins, and strikes. Environment Lock University open ground, trimmed grass field with dirt patches. White painted field lines. Overhead campus lamp posts casting pools of light. Everything outside the light pools is deep shadow. University buildings faintly visible in the background darkness. Dust and grass kick up on every impact, roll, and landing. Timeline 0.0–2.5s: Fighter A stands behind Fighter B. Right palm covers her mouth from behind. She grabs at his wrist, eyes sharp and furious. She drives her elbow into his ribs, breaks free, spins out. She lands in a fighting stance, dust rising. 2.5–5.0s: She rolls her shoulders, cracks her neck. Walks forward with slow, aggressive confidence. Speaks clearly in English to camera: "You really thought you could catch me off guard?" 5.0–7.0s: He sprints forward, throws a right hook. She ducks low, sweeps his front leg. He stumbles, catches balance, resets stance. 7.0–9.0s: He throws a straight right jab. She parries with her left forearm, steps in. Delivers a sharp right knee strike to his side. He staggers back two steps. 9.0–12.5s: He lands a glancing punch on her shoulder. She barely reacts, grabs his wrist, twists. Executes a real hip-throw, slams him to the ground. He rolls once, pushes up onto one knee, breathing hard. 12.5–16.0s: Camera pushes in on her face. Sweat, dirt, sharp determined eyes. First orange-red flame threads flicker along her fingertips. Threads crawl up her arms to her shoulders. Her eyes begin to glow amber-orange from within. The lamp posts start flickering irregularly. 16.0–18.0s: She exhales sharply, rolls her wrists. Small flames bloom across both palms. Fabric of her top singes faintly at the edges. Heat distortion ripples the air around her. 18.0–21.0s: Camera circles her fast. He rises, throws a desperate punch. She sidesteps, grabs his arm, flame licking her grip. Grass scorches in a ring beneath her feet. Flame trails follow every movement of her limbs. 21.0–23.0s: She unleashes a flurry — spinning back kick, followed by a jumping roundhouse. Each strike leaves a brief trail of fire in the air. He blocks two, takes the third across the jaw. He drops to one knee, dazed, smoke rising off his shoulder. 23.0–27.5s: She plants her feet, pulls both fists back, flames coiling tight around her forearms. Steps in with full aggression, drives a flaming knee strike upward into his chest. Impact releases a burst of fire and sparks outward. 27.5–30.0s: He's launched back, lands hard, skidding through dirt. She stands over him, flames dying down her arms to embers. Breathing hard, one fist still smoking. Lamp posts dim to near darkness. Only the fading orange glow lights her face. Then everything goes dark. Lighting Extremely dark night, deep shadows. Lamp posts as rim light on shoulders and hair. Warm orange-red glow after her power appears, casting light on both fighters. Sound Design Location sound only. Grass and dirt scraping, fabric rustling. Sharp impact sounds, heavy breathing, grunts. Flame ignition crackle building to a low roar. Lamp buzz as they flicker. Her line delivered clearly at 2.5–5.0s. Sharp exhale and crackle at the final blow. Total silence at the very end. No background music throughout.

Shami

30,581 次观看 • 15 天前