Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

(1/2) 📢📢𝗦𝗰𝗲𝗻𝗲𝗧𝗲𝘅 📢📢 Given scene geometry and text prompt -> SceneTex generates high-quality textures. Main idea: directly optimize scene texture with gradients from a score-distillation objective with view sampling.

MattNiessner's profile picture

Matthias Niessner

48,723 subscribers

40,835 Aufrufe • vor 2 Jahren •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

We present “3D magician”: TADA! Text to Animatable Digital Avatars. Given a textual description as input only, our method TADA generates expressive animatable 3D avatars with high-quality geometry and lifelike textures. (1/10)
0:14

We present “3D magician”: TADA! Text to Animatable Digital Avatars. Given a textual description as input only, our method TADA generates expressive animatable 3D avatars with high-quality geometry and lifelike textures. (1/10)

Hongwei Yi

52,306 Aufrufe • vor 3 Jahren

📢🚗✨ Excited to announce InfiniCube, our scalable generative model for dynamic 3D driving scene generation with high fidelity and controllability! InfiniCube generates very large-scale (300m×400m ~ 100,000m^2), dynamic 3D driving scenes given HD maps, 3D bounding boxes, and text prompts as controls. Website: Props to Yifan Lu for leading this project!!!! [1/N]
0:16

📢🚗✨ Excited to announce InfiniCube, our scalable generative model for dynamic 3D driving scene generation with high fidelity and controllability! InfiniCube generates very large-scale (300m×400m ~ 100,000m^2), dynamic 3D driving scenes given HD maps, 3D bounding boxes, and text prompts as controls. Website: Props to Yifan Lu for leading this project!!!! [1/N]

Xuanchi Ren

26,638 Aufrufe • vor 1 Jahr

📢We introduce “RefFusion”, a novel inpainting method for scenes reconstructed using 3D Gaussian Splatting. 🔗 TLDR: we personalize an image diffusion model to a given reference image and distill its knowledge to 3D through score distillation sampling.
0:11

📢We introduce “RefFusion”, a novel inpainting method for scenes reconstructed using 3D Gaussian Splatting. 🔗 TLDR: we personalize an image diffusion model to a given reference image and distill its knowledge to 3D through score distillation sampling.

Ashkan Mirzaei

34,700 Aufrufe • vor 2 Jahren

📢 LiteReality: Graphics-Ready 3D Scene Reconstruction from RGB-D Scans🏠✨ -> converts RGB-D scans into compact, realistic, and interactive 3D scenes — featuring high-quality meshes, PBR materials, and articulated objects. 📷 🌍
0:19

📢 LiteReality: Graphics-Ready 3D Scene Reconstruction from RGB-D Scans🏠✨ -> converts RGB-D scans into compact, realistic, and interactive 3D scenes — featuring high-quality meshes, PBR materials, and articulated objects. 📷 🌍

Matthias Niessner

23,855 Aufrufe • vor 1 Jahr

📢DNF: Generating 4D animations with dictionary-based neural fields! Xinyi Zhang presents a new dictionary-based neural field for unconditional 4D generation of deforming shapes -- generating motions with high-quality shape and temporal consistency.
0:19

📢DNF: Generating 4D animations with dictionary-based neural fields! Xinyi Zhang presents a new dictionary-based neural field for unconditional 4D generation of deforming shapes -- generating motions with high-quality shape and temporal consistency.

Angela Dai

14,643 Aufrufe • vor 1 Jahr

📢New in ScanNet++: High-Res 360° Panos! Chandan Yeshwanth & Yueh-Cheng Liu have added pano captures for 956 ScanNet++ scenes, fully aligned with the 3D meshes, DSLR, and iPhone data - multiple panos per scene Check it out: Docs Code
0:11

📢New in ScanNet++: High-Res 360° Panos! Chandan Yeshwanth & Yueh-Cheng Liu have added pano captures for 956 ScanNet++ scenes, fully aligned with the 3D meshes, DSLR, and iPhone data - multiple panos per scene Check it out: Docs Code

Angela Dai

12,896 Aufrufe • vor 10 Monaten

📢GeomHair: Reconstruction of Hair Strands from Colorless 3D Scans📢 We present a novel method to reconstruct hair strands from colorless 3D scans by extracting orientation cues directly from the mesh surface geometry by finding local characteristic lines and from shaded renderings using a neural 2D line detector. We enhance the reconstruction with a diffusion prior trained on synthetic hair data and adapted to each scan using a tailored text prompt, allowing us to recover both simple and complex hairstyles without relying on color input. To support further research, we also introduce Strands400, the largest publicly available dataset of 3D hair strand reconstructions from real-world scans of 400 different people, featuring complicated hairstyles, such as ponytails and buns. 🌍 📷 Great work by Rachmadio Noval L. Artem Sevastopolsky Egor Zakharov @ness_pris
0:15

📢GeomHair: Reconstruction of Hair Strands from Colorless 3D Scans📢 We present a novel method to reconstruct hair strands from colorless 3D scans by extracting orientation cues directly from the mesh surface geometry by finding local characteristic lines and from shaded renderings using a neural 2D line detector. We enhance the reconstruction with a diffusion prior trained on synthetic hair data and adapted to each scan using a tailored text prompt, allowing us to recover both simple and complex hairstyles without relying on color input. To support further research, we also introduce Strands400, the largest publicly available dataset of 3D hair strand reconstructions from real-world scans of 400 different people, featuring complicated hairstyles, such as ponytails and buns. 🌍 📷 Great work by Rachmadio Noval L. Artem Sevastopolsky Egor Zakharov @ness_pris

Matthias Niessner

12,499 Aufrufe • vor 1 Jahr

📢 🧵 1/ We’re thrilled to announce that LimeWire has acquired BlueWillow, a leading image generation model and the world's 2nd largest Discord community with 2+ million users. Read on to learn what this acquisition means for LimeWire, our community and the $LMWR token.
0:17

📢 🧵 1/ We’re thrilled to announce that LimeWire has acquired BlueWillow, a leading image generation model and the world's 2nd largest Discord community with 2+ million users. Read on to learn what this acquisition means for LimeWire, our community and the $LMWR token.

LimeWire

64,150 Aufrufe • vor 3 Jahren

In the next update we will add a voting feature to our staking platform We are planning to launch the first voting on Sunday and invite XRPL projects to take part in it. 📢Tag project with the strongest community and we will invite them to vote! (1/2)⬇️
0:11

In the next update we will add a voting feature to our staking platform We are planning to launch the first voting on Sunday and invite XRPL projects to take part in it. 📢Tag project with the strongest community and we will invite them to vote! (1/2)⬇️

Rocket Launch

19,610 Aufrufe • vor 3 Jahren

Aesthetic Morning Routine Start Your Day With Positive Energy Made with seedance 2.0 on WaveSpeedAI Prompt: Scene 1 — 0–3 sec: She wakes up in a cozy bedroom as warm sunlight enters through the curtains. Scene 2 — 3–6 sec: She opens the curtains, takes a fresh breath, and enjoys the morning view. Scene 3 — 6–9 sec: In the bathroom, she washes her face and applies moisturizer. Scene 4 — 9–12 sec: She brushes her hair, gets dressed, and checks herself in the mirror. Scene 5 — 12–15 sec: She grabs her bag and walks outside into the bright morning, smiling naturally. Cinematic, realistic, smooth camera movement.
0:15

Aesthetic Morning Routine Start Your Day With Positive Energy Made with seedance 2.0 on WaveSpeedAI Prompt: Scene 1 — 0–3 sec: She wakes up in a cozy bedroom as warm sunlight enters through the curtains. Scene 2 — 3–6 sec: She opens the curtains, takes a fresh breath, and enjoys the morning view. Scene 3 — 6–9 sec: In the bathroom, she washes her face and applies moisturizer. Scene 4 — 9–12 sec: She brushes her hair, gets dressed, and checks herself in the mirror. Scene 5 — 12–15 sec: She grabs her bag and walks outside into the bright morning, smiling naturally. Cinematic, realistic, smooth camera movement.

Noor

13,958 Aufrufe • vor 24 Tagen

Flawless finish, every stroke meet the ultimate brush set for your everyday glam, Created with Seedance 2.5 Prompt: High end commercial makeup ad, multi scene video sequence. Scene 1: Close-up of a blonde model with flawless skin and light makeup looking at the camera. Scene 2: Macro shot of a luxurious white-and-gold makeup brush sweeping into a compact of soft pink powder, creating a slight dust cloud. Scene 3: A red-haired model having blusher applied to her cheek with a brush in a professional studio setup. Scene 4: Precision eye makeup application, applying golden eyeshadow with a fine brush. Scene 5: Close-up of glossy dark-pink lips being perfected with a lip brush. Scene 6: A brush highlighting the collarbone with shimmer powder. Scene 7: Close-up portrait of an Asian model with elegant dark hair and polished makeup. Scene 8: Three diverse models (Asian, blonde, and red-haired) wearing stylish professional business suits and white pants, confidently walking forward holding makeup brushes. Scene 9: Product showcase of the "adelice" white-and-gold makeup brush set neatly arranged with containers and compacts on a reflective glass surface. Cinematic studio lighting, photorealistic, 8k resolution, smooth camera pans and transitions.
0:18

Flawless finish, every stroke meet the ultimate brush set for your everyday glam, Created with Seedance 2.5 Prompt: High end commercial makeup ad, multi scene video sequence. Scene 1: Close-up of a blonde model with flawless skin and light makeup looking at the camera. Scene 2: Macro shot of a luxurious white-and-gold makeup brush sweeping into a compact of soft pink powder, creating a slight dust cloud. Scene 3: A red-haired model having blusher applied to her cheek with a brush in a professional studio setup. Scene 4: Precision eye makeup application, applying golden eyeshadow with a fine brush. Scene 5: Close-up of glossy dark-pink lips being perfected with a lip brush. Scene 6: A brush highlighting the collarbone with shimmer powder. Scene 7: Close-up portrait of an Asian model with elegant dark hair and polished makeup. Scene 8: Three diverse models (Asian, blonde, and red-haired) wearing stylish professional business suits and white pants, confidently walking forward holding makeup brushes. Scene 9: Product showcase of the "adelice" white-and-gold makeup brush set neatly arranged with containers and compacts on a reflective glass surface. Cinematic studio lighting, photorealistic, 8k resolution, smooth camera pans and transitions.

Avelyrah

14,843 Aufrufe • vor 1 Tag

#AI x #Metaverse - #XANA is venturing into the world's most cutting-edge metaverse powered by AI. Introducing Phase 1: 'Text to Skybox' released today. With just a prompt, anyone can easily create high-quality spaces outside their world, elevating the world experience to another dimension. Stay tuned for Phases 2 and 3 of these AI x Metaverse features!
0:20

#AI x #Metaverse - #XANA is venturing into the world's most cutting-edge metaverse powered by AI. Introducing Phase 1: 'Text to Skybox' released today. With just a prompt, anyone can easily create high-quality spaces outside their world, elevating the world experience to another dimension. Stay tuned for Phases 2 and 3 of these AI x Metaverse features!

XANA

23,690 Aufrufe • vor 3 Jahren

Nano banana pro on Gemini+ Seedance 2.0 Prompt: A high-end UGC fashion campaign featuring a young East Asian woman with long, sleek straight black hair, showcasing a product. She is wearing a deep maroon knit cowl-neck short-sleeve top with elegant draped pleats on the shoulder and ruching on the sides, paired with white drawstring cargo shorts. A multi-angle cinematic sequence style, showcasing both full-body lifestyle shots and close-up fabric textures. The setting is a minimalist, modern aesthetic apartment with a cream-colored curved sofa, a large gold-rimmed standing mirror, and bright windows. Soft, diffused natural morning sunlight illuminating the scene, casting gentle shadows. Photorealistic, high-end commercial quality, shot on an 85mm lens, f/1.8 aperture, sharp focus on the fine texture of the burgundy fabric, clean color grading, 8K resolution.
0:10

Nano banana pro on Gemini+ Seedance 2.0 Prompt: A high-end UGC fashion campaign featuring a young East Asian woman with long, sleek straight black hair, showcasing a product. She is wearing a deep maroon knit cowl-neck short-sleeve top with elegant draped pleats on the shoulder and ruching on the sides, paired with white drawstring cargo shorts. A multi-angle cinematic sequence style, showcasing both full-body lifestyle shots and close-up fabric textures. The setting is a minimalist, modern aesthetic apartment with a cream-colored curved sofa, a large gold-rimmed standing mirror, and bright windows. Soft, diffused natural morning sunlight illuminating the scene, casting gentle shadows. Photorealistic, high-end commercial quality, shot on an 85mm lens, f/1.8 aperture, sharp focus on the fine texture of the burgundy fabric, clean color grading, 8K resolution.

Calira

12,671 Aufrufe • vor 3 Monaten

NVIDIA DROPPED A MOTION DIFFUSION MODEL FOR HUMANOID ROBOTS trained on 700 hours of mocap data kimodo generates high-quality 3D human and robot motions from text prompts you control it with: → full-body pose keyframes → end-effector positions/rotations → 2D paths and waypoints works on human skeletons and unitree G1 robot plug the outputs directly into mujoco or retarget to other robots using GMR has a web-based interactive demo with a timeline editor. runs locally needs ~17GB VRAM to run inference open source under apache 2.0
0:11

NVIDIA DROPPED A MOTION DIFFUSION MODEL FOR HUMANOID ROBOTS trained on 700 hours of mocap data kimodo generates high-quality 3D human and robot motions from text prompts you control it with: → full-body pose keyframes → end-effector positions/rotations → 2D paths and waypoints works on human skeletons and unitree G1 robot plug the outputs directly into mujoco or retarget to other robots using GMR has a web-based interactive demo with a timeline editor. runs locally needs ~17GB VRAM to run inference open source under apache 2.0

Vaishnavi

17,572 Aufrufe • vor 4 Monaten

📢Pix2NPHM: Learning to Regress NPHM Reconstructions From a Single Image📢 We directly regress neural parametric head models (NPHMs) from a single image — fast, stable, and significantly more expressive than classical 3DMMs such as FLAME. Face tracking & 3D reconstruction are often limited by the representational capacity of PCA-based face models. By lifting NPHMs to a first-class reconstruction primitive, we enable more accurate geometry, richer expressions, and finer animation control. Pix2NPHM obtains fast and reliable NPHM reconstructions on real-world data. Inference-time optimization against surface normals and canonical point maps can further increase fidelity. Key to successful and generalized training of our ViT-based network are: (1) large-scale registration of existing 3D head datasets, and (2) self-supervised training on vast in-the-wild 2D video datasets using pseudo ground-truth surface normals. Finally, we show that geometry-aware pretraining on pixel-aligned reconstruction tasks significantly outperforms generic visual pretraining (e.g., DINO-style features) in terms of generalization. 🌍 🎥 Great work by Simon Giebenhain, Tobias Kirschstein, Liam Schoneveld, Davide Davoli, Zhe Chen
0:17

📢Pix2NPHM: Learning to Regress NPHM Reconstructions From a Single Image📢 We directly regress neural parametric head models (NPHMs) from a single image — fast, stable, and significantly more expressive than classical 3DMMs such as FLAME. Face tracking & 3D reconstruction are often limited by the representational capacity of PCA-based face models. By lifting NPHMs to a first-class reconstruction primitive, we enable more accurate geometry, richer expressions, and finer animation control. Pix2NPHM obtains fast and reliable NPHM reconstructions on real-world data. Inference-time optimization against surface normals and canonical point maps can further increase fidelity. Key to successful and generalized training of our ViT-based network are: (1) large-scale registration of existing 3D head datasets, and (2) self-supervised training on vast in-the-wild 2D video datasets using pseudo ground-truth surface normals. Finally, we show that geometry-aware pretraining on pixel-aligned reconstruction tasks significantly outperforms generic visual pretraining (e.g., DINO-style features) in terms of generalization. 🌍 🎥 Great work by Simon Giebenhain, Tobias Kirschstein, Liam Schoneveld, Davide Davoli, Zhe Chen

Matthias Niessner

37,965 Aufrufe • vor 8 Monaten

Exploring the Beauty of the World Through an Unforgettable 15-Second Journey Made with seedance 2.0 on WaveSpeedAI Prompt: A beautiful young female travel vlogger exploring a different destination in every scene, energetic and joyful, cinematic travel documentary style, ultra-realistic, 8K, HDR, natural lighting, smooth camera movements, vibrant colors, premium travel film aesthetic, photorealistic, masterpiece. Scene 1 (0–2s): Walking through a modern airport with a suitcase and backpack, smiling at the camera, ready for adventure. Scene 2 (2–4s): Riding a scenic train through lush green mountains, filming the breathtaking view from the window. Scene 3 (4–6s): Standing on a mountain viewpoint at sunrise, arms open, capturing the landscape with a vlog camera. Scene 4 (6–8s): Walking barefoot along a tropical beach, waves touching the shore, flying drone shot. Scene 5 (8–10s): Exploring colorful streets and a local market, tasting street food and interacting with locals. Scene 6 (10–12s): Visiting a majestic waterfall surrounded by a lush forest, cinematic slow-motion shots. Scene 7 (12–15s): Standing on a rooftop overlooking a glowing city skyline at sunset, waving at the camera with the text "Where should I travel next?", ending with a smooth cinematic fade-out. Style: Hyper-realistic, luxury travel vlog, smooth transitions, dynamic camera movement, cinematic color grading, volumetric lighting, realistic facial expressions, premium YouTube travel content, highly detailed, viral travel reel, masterpiece.
0:15

Exploring the Beauty of the World Through an Unforgettable 15-Second Journey Made with seedance 2.0 on WaveSpeedAI Prompt: A beautiful young female travel vlogger exploring a different destination in every scene, energetic and joyful, cinematic travel documentary style, ultra-realistic, 8K, HDR, natural lighting, smooth camera movements, vibrant colors, premium travel film aesthetic, photorealistic, masterpiece. Scene 1 (0–2s): Walking through a modern airport with a suitcase and backpack, smiling at the camera, ready for adventure. Scene 2 (2–4s): Riding a scenic train through lush green mountains, filming the breathtaking view from the window. Scene 3 (4–6s): Standing on a mountain viewpoint at sunrise, arms open, capturing the landscape with a vlog camera. Scene 4 (6–8s): Walking barefoot along a tropical beach, waves touching the shore, flying drone shot. Scene 5 (8–10s): Exploring colorful streets and a local market, tasting street food and interacting with locals. Scene 6 (10–12s): Visiting a majestic waterfall surrounded by a lush forest, cinematic slow-motion shots. Scene 7 (12–15s): Standing on a rooftop overlooking a glowing city skyline at sunset, waving at the camera with the text "Where should I travel next?", ending with a smooth cinematic fade-out. Style: Hyper-realistic, luxury travel vlog, smooth transitions, dynamic camera movement, cinematic color grading, volumetric lighting, realistic facial expressions, premium YouTube travel content, highly detailed, viral travel reel, masterpiece.

Noor

16,700 Aufrufe • vor 1 Monat

Have you ever imagined your city turning into a Lego world? With Kling O1, that fantasy becomes real! Tools: Kling IMAGE O1, Kling Image to Video Use this image-generation prompt: Present a clear, 45° top-down isometric miniature Lego-style 3D cartoon scene of New York built entirely from plastic bricks. Recreate the city’s most iconic landmarks and architectural elements using distinctive Lego block geometry. Use clean, glossy brick textures with subtle PBR reflections and soft, playful lighting and shadows. Keep the composition minimalistic with a soft, solid-colored background. Feeling inspired? Just change the city name and watch a whole new Lego universe come to life!
0:11

Have you ever imagined your city turning into a Lego world? With Kling O1, that fantasy becomes real! Tools: Kling IMAGE O1, Kling Image to Video Use this image-generation prompt: Present a clear, 45° top-down isometric miniature Lego-style 3D cartoon scene of New York built entirely from plastic bricks. Recreate the city’s most iconic landmarks and architectural elements using distinctive Lego block geometry. Use clean, glossy brick textures with subtle PBR reflections and soft, playful lighting and shadows. Keep the composition minimalistic with a soft, solid-colored background. Feeling inspired? Just change the city name and watch a whole new Lego universe come to life!

Kling AI

38,391 Aufrufe • vor 8 Monaten

GM my friends!🌞 Are you ready to fight?!😏🔥 Megan Fox, Kat Dennings 💪🏻 👉🏻Subscribe for more content!😎 Nano Banana 2 viao Hailuo AI & Kling 3.0 via Higgsfield AI Prompt: Hyper-realistic photorealistic portrait of cosplay Megam fox as Chun-Li from Street Fighter, crouching pose with hands on the ground, looking slightly directly to the camera with a neutral expression, extremely detailed face, 8k resolution, photorealistic skin texture with visible pores, subtle skin imperfections, realistic sweat droplets, micro skin details, individual eyelashes, realistic brown eyes with detailed iris texture and catchlights, natural subsurface scattering, ultra-detailed black hair with individual strands and realistic highlights, white silk ribbons in hair.She wears the classic blue qipao dress with gold embroidery, white high heels stilettos, black spiked leather wristbands with metallic studs, highly detailed fabric textures, realistic cloth folds and wrinkles, reflective surfaces on the dress. Torn paper collage effect in the background with two close-up realistic eye photos (one left, one right) overlapping the main image.Shot on Sony A1 camera with Sony FE 85mm f/1.4 GM lens, f/2.0 aperture, 1/160s shutter speed, ISO 100, cinematic natural studio lighting, soft key light from top left, subtle rim lighting, dramatic yet clean white background, sharp focus on face and body, shallow depth of field, bokeh effect, ultra-detailed 32k, masterpiece, best quality, photorealism, national geographic level detail, 8k raw photo, hyperrealistic textures, microscopic skin detail, octane render, unreal engine 5 quality. "style_and_quality": { "style": "photorealistic, high-fashion glamour photography, cinematic", "detail_level": "ultra high detail on skin pores, fabric texture, hair strands, reflections", "rendering": "realistic fabric details, glossy materials, accurate light interaction", "avoid": "cartoon, anime, illustration, painting, 3D render look" } }
0:15

GM my friends!🌞 Are you ready to fight?!😏🔥 Megan Fox, Kat Dennings 💪🏻 👉🏻Subscribe for more content!😎 Nano Banana 2 viao Hailuo AI & Kling 3.0 via Higgsfield AI Prompt: Hyper-realistic photorealistic portrait of cosplay Megam fox as Chun-Li from Street Fighter, crouching pose with hands on the ground, looking slightly directly to the camera with a neutral expression, extremely detailed face, 8k resolution, photorealistic skin texture with visible pores, subtle skin imperfections, realistic sweat droplets, micro skin details, individual eyelashes, realistic brown eyes with detailed iris texture and catchlights, natural subsurface scattering, ultra-detailed black hair with individual strands and realistic highlights, white silk ribbons in hair.She wears the classic blue qipao dress with gold embroidery, white high heels stilettos, black spiked leather wristbands with metallic studs, highly detailed fabric textures, realistic cloth folds and wrinkles, reflective surfaces on the dress. Torn paper collage effect in the background with two close-up realistic eye photos (one left, one right) overlapping the main image.Shot on Sony A1 camera with Sony FE 85mm f/1.4 GM lens, f/2.0 aperture, 1/160s shutter speed, ISO 100, cinematic natural studio lighting, soft key light from top left, subtle rim lighting, dramatic yet clean white background, sharp focus on face and body, shallow depth of field, bokeh effect, ultra-detailed 32k, masterpiece, best quality, photorealism, national geographic level detail, 8k raw photo, hyperrealistic textures, microscopic skin detail, octane render, unreal engine 5 quality. "style_and_quality": { "style": "photorealistic, high-fashion glamour photography, cinematic", "detail_level": "ultra high detail on skin pores, fabric texture, hair strands, reflections", "rendering": "realistic fabric details, glossy materials, accurate light interaction", "avoid": "cartoon, anime, illustration, painting, 3D render look" } }

KeorUnreal

31,169 Aufrufe • vor 2 Monaten