Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

🤔🤔Tired of static, lifeless image edits? Not anymore! 🤗 🚀🚀We introduce MotionEdit, a framework supporting image editing that understands action, motion, interaction beyond static changes! 🤩🤩 🔗Full paper: ✨Project page:

yixin_wan_'s profile picture

Yixin Wan @ACL2026

2,925 subscribers

59,481 Aufrufe • vor 8 Monaten •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

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.
0:46

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.

AK

71,232 Aufrufe • vor 1 Jahr

TokenFlow: Consistent Diffusion Features for Consistent Video Editing paper page: The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos.
0:36

TokenFlow: Consistent Diffusion Features for Consistent Video Editing paper page: The generative AI revolution has recently expanded to videos. Nevertheless, current state-of-the-art video models are still lagging behind image models in terms of visual quality and user control over the generated content. In this work, we present a framework that harnesses the power of a text-to-image diffusion model for the task of text-driven video editing. Specifically, given a source video and a target text-prompt, our method generates a high-quality video that adheres to the target text, while preserving the spatial layout and motion of the input video. Our method is based on a key observation that consistency in the edited video can be obtained by enforcing consistency in the diffusion feature space. We achieve this by explicitly propagating diffusion features based on inter-frame correspondences, readily available in the model. Thus, our framework does not require any training or fine-tuning, and can work in conjunction with any off-the-shelf text-to-image editing method. We demonstrate state-of-the-art editing results on a variety of real-world videos.

AK

143,469 Aufrufe • vor 3 Jahren

🚀 Introducing InterDyn — our newly accepted CVPR work that explores controllable synthesis of interactive dynamics! Building upon powerful video diffusion models, InterDyn infers future motion and interactions directly from an input image and a dynamic control signal (e.g., a moving hand mask). Check out how we push the boundaries of intuitive physics with video generative models. Project page: Arxiv: #GenAI #AIGC #VideoGen #ML #ComputerVision #CVPR2025 🧵1/6
0:28

🚀 Introducing InterDyn — our newly accepted CVPR work that explores controllable synthesis of interactive dynamics! Building upon powerful video diffusion models, InterDyn infers future motion and interactions directly from an input image and a dynamic control signal (e.g., a moving hand mask). Check out how we push the boundaries of intuitive physics with video generative models. Project page: Arxiv: #GenAI #AIGC #VideoGen #ML #ComputerVision #CVPR2025 🧵1/6

Haven (Haiwen) Feng

44,898 Aufrufe • vor 1 Jahr

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):
1:12

Today, we introduce HunyuanImage 3.0-Instruct, a native multimodal model focusing on image-editing by integrating visual understanding with precise image synthesis! 🚀 It understands input images and reasons before generating images. Built on an 80B-parameter MoE architecture (13B activated), it natively unifies deep multimodal comprehension and high-fidelity generation. 🧠 A "Thinking" Model with Native CoT & MixGRPO: The model doesn’t just execute commands, it processes them through a Native Chain-of-Thought (CoT) schema. Enhanced by our self-developed MixGRPO algorithm, it reasons through complex instructions to achieve flawless intent alignment and human-preference consistency. 🎨 Precise Editing & Multi-Image Fusion: The model enables accurate image editing by adding, removing, or modifying elements while keeping non-target areas perfectly intact. It also excels at seamless multi-image fusion, synthesizing complex scenes by extracting and blending elements from multiple sources into a unified, consistent output. 🏆 SOTA Performance: HunyuanImage 3.0-Instruct sets a new benchmark in visual quality and alignment, delivering performance that matches leading proprietary models. We aim to enable the community to explore new ideas with a state-of-the-art foundation model, fostering a dynamic and vibrant image generation ecosystem. 🛠️🎨 💻Try it at (PC only):

Tencent Hy

125,803 Aufrufe • vor 6 Monaten

VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams paper page: Neural Radiance Fields (NeRFs) excel in photorealistically rendering static scenes. However, rendering dynamic, long-duration radiance fields on ubiquitous devices remains challenging, due to data storage and computational constraints. In this paper, we introduce VideoRF, the first approach to enable real-time streaming and rendering of dynamic radiance fields on mobile platforms. At the core is a serialized 2D feature image stream representing the 4D radiance field all in one. We introduce a tailored training scheme directly applied to this 2D domain to impose the temporal and spatial redundancy of the feature image stream. By leveraging the redundancy, we show that the feature image stream can be efficiently compressed by 2D video codecs, which allows us to exploit video hardware accelerators to achieve real-time decoding. On the other hand, based on the feature image stream, we propose a novel rendering pipeline for VideoRF, which has specialized space mappings to query radiance properties efficiently. Paired with a deferred shading model, VideoRF has the capability of real-time rendering on mobile devices thanks to its efficiency. We have developed a real-time interactive player that enables online streaming and rendering of dynamic scenes, offering a seamless and immersive free-viewpoint experience across a range of devices, from desktops to mobile phones.
4:56

VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams paper page: Neural Radiance Fields (NeRFs) excel in photorealistically rendering static scenes. However, rendering dynamic, long-duration radiance fields on ubiquitous devices remains challenging, due to data storage and computational constraints. In this paper, we introduce VideoRF, the first approach to enable real-time streaming and rendering of dynamic radiance fields on mobile platforms. At the core is a serialized 2D feature image stream representing the 4D radiance field all in one. We introduce a tailored training scheme directly applied to this 2D domain to impose the temporal and spatial redundancy of the feature image stream. By leveraging the redundancy, we show that the feature image stream can be efficiently compressed by 2D video codecs, which allows us to exploit video hardware accelerators to achieve real-time decoding. On the other hand, based on the feature image stream, we propose a novel rendering pipeline for VideoRF, which has specialized space mappings to query radiance properties efficiently. Paired with a deferred shading model, VideoRF has the capability of real-time rendering on mobile devices thanks to its efficiency. We have developed a real-time interactive player that enables online streaming and rendering of dynamic scenes, offering a seamless and immersive free-viewpoint experience across a range of devices, from desktops to mobile phones.

AK

38,686 Aufrufe • vor 2 Jahren

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.
0:21

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

153,241 Aufrufe • vor 3 Jahren

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.
1:43

Google presents Still-Moving Customized Video Generation without Customized Video Data Customizing text-to-image (T2I) models has seen tremendous progress recently, particularly in areas such as personalization, stylization, and conditional generation. However, expanding this progress to video generation is still in its infancy, primarily due to the lack of customized video data. In this work, we introduce Still-Moving, a novel generic framework for customizing a text-to-video (T2V) model, without requiring any customized video data. The framework applies to the prominent T2V design where the video model is built over a text-to-image (T2I) model (e.g., via inflation). We assume access to a customized version of the T2I model, trained only on still image data (e.g., using DreamBooth or StyleDrop). Naively plugging in the weights of the customized T2I model into the T2V model often leads to significant artifacts or insufficient adherence to the customization data. To overcome this issue, we train lightweight Spatial Adapters that adjust the features produced by the injected T2I layers. Importantly, our adapters are trained on "frozen videos" (i.e., repeated images), constructed from image samples generated by the customized T2I model. This training is facilitated by a novel Motion Adapter module, which allows us to train on such static videos while preserving the motion prior of the video model. At test time, we remove the Motion Adapter modules and leave in only the trained Spatial Adapters. This restores the motion prior of the T2V model while adhering to the spatial prior of the customized T2I model. We demonstrate the effectiveness of our approach on diverse tasks including personalized, stylized, and conditional generation. In all evaluated scenarios, our method seamlessly integrates the spatial prior of the customized T2I model with a motion prior supplied by the T2V model.

AK

40,485 Aufrufe • vor 2 Jahren

We’re excited to share DiT4DiT, an end-to-end Video-Action Model for robot learning that unifies a video Diffusion Transformer and an action Diffusion Transformer in a single cascaded framework. By leveraging the rich spatiotemporal and physical dynamics learned through video generation, rather than static image-text priors, DiT4DiT achieves state-of-the-art results on LIBERO (98.6%) and RoboCasa GR1 (50.8%) with far less training data, delivering over 10× better sample efficiency and up to 7× faster convergence. Real-world deployment on a humanoid robot further shows robust generalization. We believe this is a step toward making video generation a powerful backbone for robot policy learning. This work builds upon the brilliant foundations laid by Nvidia's GR00T and Cosmos. Project: Paper: Code: Coming soon. In the meantime, you can ask your coding agent to reproduce the method based on GR00T/Cosmos.
2:32

We’re excited to share DiT4DiT, an end-to-end Video-Action Model for robot learning that unifies a video Diffusion Transformer and an action Diffusion Transformer in a single cascaded framework. By leveraging the rich spatiotemporal and physical dynamics learned through video generation, rather than static image-text priors, DiT4DiT achieves state-of-the-art results on LIBERO (98.6%) and RoboCasa GR1 (50.8%) with far less training data, delivering over 10× better sample efficiency and up to 7× faster convergence. Real-world deployment on a humanoid robot further shows robust generalization. We believe this is a step toward making video generation a powerful backbone for robot policy learning. This work builds upon the brilliant foundations laid by Nvidia's GR00T and Cosmos. Project: Paper: Code: Coming soon. In the meantime, you can ask your coding agent to reproduce the method based on GR00T/Cosmos.

Shuo Yang

31,872 Aufrufe • vor 5 Monaten

If you want a vision encoder for dexterous manipulation, what should be the most important part to model? 🤔 Current standard models like CLIP, SigLIP, and DINOv2 have an incredible grasp of semantics and spatial details. But they lack the action-centric structure needed for downstream visuomotor control. But collecting annotated robotic trajectories at scale is SUPER expensive and largely unrealistic. So, how do we bridge this gap? We introduce CAIP (Contrastive Action-Image Pre-training) ⬇️ 🔸 Action-centric upstream: we align visual observations with action chunks through a contrastive objective. 🔸 Human video as a proxy: we represent 3D human hand poses analogously to robotic end-effector actions, tapping into a massive source of human demonstrations. 🔸 Massive scale: pre-trained on over 32,000 hours of manipulation video, driving both sample efficiency and robust generalization. 🔸 Hardware proven: achieves a 76% average success rate on a real-world Dexmate Vega bimanual Dexmate manipulator with dual 22-DoF Sharpa Wave hands Sharpa . 🔸 State-of-the-art: significantly outperforms strong baselines like DINOv2, SigLIP, MVP, and Qwen3.5 ViT across complex tasks, even under unexpected lighting changes and visual distractors. 🌐 Project: 📝 Blog: 📄 Paper: 💻 Model:
1:30

If you want a vision encoder for dexterous manipulation, what should be the most important part to model? 🤔 Current standard models like CLIP, SigLIP, and DINOv2 have an incredible grasp of semantics and spatial details. But they lack the action-centric structure needed for downstream visuomotor control. But collecting annotated robotic trajectories at scale is SUPER expensive and largely unrealistic. So, how do we bridge this gap? We introduce CAIP (Contrastive Action-Image Pre-training) ⬇️ 🔸 Action-centric upstream: we align visual observations with action chunks through a contrastive objective. 🔸 Human video as a proxy: we represent 3D human hand poses analogously to robotic end-effector actions, tapping into a massive source of human demonstrations. 🔸 Massive scale: pre-trained on over 32,000 hours of manipulation video, driving both sample efficiency and robust generalization. 🔸 Hardware proven: achieves a 76% average success rate on a real-world Dexmate Vega bimanual Dexmate manipulator with dual 22-DoF Sharpa Wave hands Sharpa . 🔸 State-of-the-art: significantly outperforms strong baselines like DINOv2, SigLIP, MVP, and Qwen3.5 ViT across complex tasks, even under unexpected lighting changes and visual distractors. 🌐 Project: 📝 Blog: 📄 Paper: 💻 Model:

Roei Herzig

23,239 Aufrufe • vor 1 Monat

🚀 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)
0:43

🚀 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 Aufrufe • vor 1 Jahr

Aigents! 🤖 Have you ever imagined a world where technology could not only see, but also understand and advise you? 🤔 Imagine playing a game of chess, strategizing each move, and then, with the help of AigentX, discovering the next winning move. This is not just a possibility anymore, it's reality! 🚀 Introducing AigentX's Image Processing feature! This isn't just about analysing and processing images; it's about bringing them to life with insights and guidance that we once thought could never become a reality and merely a figment of our imagination! 🤯 Picture this: You're deeply engrossed in a chess match, contemplating your next move. You snap a picture of the board and send it to AigentX. In moments, you're not just seeing the pieces – you're understanding the potential for your next move! 📸 AigentX doesn't just view the image; it interprets it, offering you useful insights and it isn't just chess! This feature can be taken straight to the real world with features like chart analysis and more! 📈 But that's just the beginning. AigentX is ready to transform how you interact with any image. From art analysis to real-world problem-solving, the possibilities are endless 📈 Check out our latest video below to see AigentX in action. This is more than an update; it's a leap into a future where your interaction with images is redefined 📻 Test out AigentX for yourself and witness how it's changing the game, one image at a time. Stay tuned for more – the journey with AigentX has just begun!
1:21

Aigents! 🤖 Have you ever imagined a world where technology could not only see, but also understand and advise you? 🤔 Imagine playing a game of chess, strategizing each move, and then, with the help of AigentX, discovering the next winning move. This is not just a possibility anymore, it's reality! 🚀 Introducing AigentX's Image Processing feature! This isn't just about analysing and processing images; it's about bringing them to life with insights and guidance that we once thought could never become a reality and merely a figment of our imagination! 🤯 Picture this: You're deeply engrossed in a chess match, contemplating your next move. You snap a picture of the board and send it to AigentX. In moments, you're not just seeing the pieces – you're understanding the potential for your next move! 📸 AigentX doesn't just view the image; it interprets it, offering you useful insights and it isn't just chess! This feature can be taken straight to the real world with features like chart analysis and more! 📈 But that's just the beginning. AigentX is ready to transform how you interact with any image. From art analysis to real-world problem-solving, the possibilities are endless 📈 Check out our latest video below to see AigentX in action. This is more than an update; it's a leap into a future where your interaction with images is redefined 📻 Test out AigentX for yourself and witness how it's changing the game, one image at a time. Stay tuned for more – the journey with AigentX has just begun!

AGIX | $AGX

22,621 Aufrufe • vor 2 Jahren

🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives. Key Advantages: 🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow. 🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement. 🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses inference steps, boosting speed and lowering costs. This allows the quantized 13B model to run on consumer-grade GPUs like the RTX 4090. Project Page: Code: Technical Report: Hugging Face：
0:40

🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives. Key Advantages: 🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow. 🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement. 🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses inference steps, boosting speed and lowering costs. This allows the quantized 13B model to run on consumer-grade GPUs like the RTX 4090. Project Page: Code: Technical Report: Hugging Face：

Tencent Hy

169,757 Aufrufe • vor 1 Jahr

How do you give a humanoid the general motion capability? Not just single motions, but all motion? Introducing SONIC, our new work on supersizing motion tracking for natural humanoid control. We argue that motion tracking is the scalable foundation task for humanoids. So we "supersized" it: 9k+ GPU hours and 100M+ motion frames. But tracking alone is not enough; we show how to make a useful control system out of it: - Universal Kinematic Planner: Enables game-like gamepad control and high-level teleoperation, just like controlling a character in a game. - VR Full-Body Teleop: Direct, real-time whole-body control by a human wearing a VR headset. - VR Keypoint Teleop: Control the upper body (hands/head) while our planner handles robust locomotion automatically. - VLA Integration: We connect this motion tracker to autonomous Visual-Language-Action (VLA) models for autonomous task execution! We use a Universal Token Space to UNIFY this command space, turning our robust tracker into a general-purpose, programmable humanoid brain. This is the generalist "System 1" for humanoids. 🚀 Project: #Humanoids #Robotics #AI #FoundationModels #NVIDIAResearch 🧠🔥
1:32

How do you give a humanoid the general motion capability? Not just single motions, but all motion? Introducing SONIC, our new work on supersizing motion tracking for natural humanoid control. We argue that motion tracking is the scalable foundation task for humanoids. So we "supersized" it: 9k+ GPU hours and 100M+ motion frames. But tracking alone is not enough; we show how to make a useful control system out of it: - Universal Kinematic Planner: Enables game-like gamepad control and high-level teleoperation, just like controlling a character in a game. - VR Full-Body Teleop: Direct, real-time whole-body control by a human wearing a VR headset. - VR Keypoint Teleop: Control the upper body (hands/head) while our planner handles robust locomotion automatically. - VLA Integration: We connect this motion tracker to autonomous Visual-Language-Action (VLA) models for autonomous task execution! We use a Universal Token Space to UNIFY this command space, turning our robust tracker into a general-purpose, programmable humanoid brain. This is the generalist "System 1" for humanoids. 🚀 Project: #Humanoids #Robotics #AI #FoundationModels #NVIDIAResearch 🧠🔥

Zhengyi “Zen” Luo

62,326 Aufrufe • vor 9 Monaten

🚀🚀🚀Introducing HY World 1.5 (WorldPlay)! We have now open-sourced the most systemized, comprehensive real-time world model framework in the industry. In HY World 1.5, we develop WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. You can generate and explore 3D worlds simply by inputting text or images. Walk, look around, and interact like you're playing a game. Highlights: 🔹Real-Time: Generates long-horizon streaming video at 24 FPS with superior consistency. 🔹Geometric Consistency: Achieved using a Reconstituted Context Memory mechanism to dynamically rebuild context from past frames to alleviate memory attenuation 🔹Robust Control: Uses a Dual Action Representation for robust response to user keyboard and mouse inputs. 🔹Versatile Applications: Supports both first-person and third-person perspectives, enabling applications like promptable events and infinite world extension. 👉🏻Try it now: 🌐Project Page: 🔗Github: 🤗Hugging Face: 📄Technical Report:
1:06

🚀🚀🚀Introducing HY World 1.5 (WorldPlay)! We have now open-sourced the most systemized, comprehensive real-time world model framework in the industry. In HY World 1.5, we develop WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. You can generate and explore 3D worlds simply by inputting text or images. Walk, look around, and interact like you're playing a game. Highlights: 🔹Real-Time: Generates long-horizon streaming video at 24 FPS with superior consistency. 🔹Geometric Consistency: Achieved using a Reconstituted Context Memory mechanism to dynamically rebuild context from past frames to alleviate memory attenuation 🔹Robust Control: Uses a Dual Action Representation for robust response to user keyboard and mouse inputs. 🔹Versatile Applications: Supports both first-person and third-person perspectives, enabling applications like promptable events and infinite world extension. 👉🏻Try it now: 🌐Project Page: 🔗Github: 🤗Hugging Face: 📄Technical Report:

Tencent Hy

189,802 Aufrufe • vor 8 Monaten

Tencent announces AppAgent Multimodal Agents as Smartphone Users paper page: Recent advancements in large language models (LLMs) have led to the creation of intelligent agents capable of performing complex tasks. This paper introduces a novel LLM-based multimodal agent framework designed to operate smartphone applications. Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps. Central to our agent's functionality is its innovative learning method. The agent learns to navigate and use new apps either through autonomous exploration or by observing human demonstrations. This process generates a knowledge base that the agent refers to for executing complex tasks across different applications. To demonstrate the practicality of our agent, we conducted extensive testing over 50 tasks in 10 different applications, including social media, email, maps, shopping, and sophisticated image editing tools. The results affirm our agent's proficiency in handling a diverse array of high-level tasks.
1:57

Tencent announces AppAgent Multimodal Agents as Smartphone Users paper page: Recent advancements in large language models (LLMs) have led to the creation of intelligent agents capable of performing complex tasks. This paper introduces a novel LLM-based multimodal agent framework designed to operate smartphone applications. Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps. Central to our agent's functionality is its innovative learning method. The agent learns to navigate and use new apps either through autonomous exploration or by observing human demonstrations. This process generates a knowledge base that the agent refers to for executing complex tasks across different applications. To demonstrate the practicality of our agent, we conducted extensive testing over 50 tasks in 10 different applications, including social media, email, maps, shopping, and sophisticated image editing tools. The results affirm our agent's proficiency in handling a diverse array of high-level tasks.

AK

343,919 Aufrufe • vor 2 Jahren

2D is what the camera sees.👀 3D is what the world is. 🌍 Today, we’re proudly open-sourcing QuerySplat — an open-source feed-forward 3D model that turns a few unposed images into a navigable 3D scene in seconds (run on 4090🪶). Your input: a few unposed images Our output: a sharp, navigable 3D scene — generated in seconds! 🚀 🧠 organize scenes with 3D queries, not pixel-bound Gaussians 📷 no camera poses or manual calibration required ⚡ reconstruct new scenes in a single forward pass ✨ preserve sharp geometry and high-frequency appearance 🏆 achieve state-of-the-art results on DL3DV-Evaluation 🔓 code and model weights are now open-source From pixels to scenes, we are making 3D generation scalable. Try our free iOS app — it’s a lot of fun! 😆 🔗 Try it on the Web: 🔗 Paper: 🔗 Code & weights: 🔗 Project page: 🔗 #AIGC #3DGS #AI #Computervision #3D #artist #3dartist #3d
1:13

2D is what the camera sees.👀 3D is what the world is. 🌍 Today, we’re proudly open-sourcing QuerySplat — an open-source feed-forward 3D model that turns a few unposed images into a navigable 3D scene in seconds (run on 4090🪶). Your input: a few unposed images Our output: a sharp, navigable 3D scene — generated in seconds! 🚀 🧠 organize scenes with 3D queries, not pixel-bound Gaussians 📷 no camera poses or manual calibration required ⚡ reconstruct new scenes in a single forward pass ✨ preserve sharp geometry and high-frequency appearance 🏆 achieve state-of-the-art results on DL3DV-Evaluation 🔓 code and model weights are now open-source From pixels to scenes, we are making 3D generation scalable. Try our free iOS app — it’s a lot of fun! 😆 🔗 Try it on the Web: 🔗 Paper: 🔗 Code & weights: 🔗 Project page: 🔗 #AIGC #3DGS #AI #Computervision #3D #artist #3dartist #3d

InSpatio｜影溯

29,033 Aufrufe • vor 12 Tagen

One of the workflows I helped advance and lead for Destiny 2 was our static compositing pipeline for marketing visuals. When we were creating marketing content for #Destiny2, we always needed strong key visuals for weapons, characters, abilities, and other in-game elements. Unlike a traditional gameplay screenshot, where everything is baked into one static image, composited shots let us isolate individual elements and light them more precisely. That gave us a lot more flexibility to build scenes with better legibility, clearer focus, stronger lighting, improved contrast, and a more intentional overall look. One of the biggest strengths of compositing is having access to individual layers in Photoshop. That layered approach not only makes the editing process easier, but also makes the final artwork much more flexible. Images can be adapted for trailer motion graphics, reworked for print, or pushed further as creative needs evolve. Destiny 2 composites are built using Adobe Photoshop alongside in-game captures. Using advanced development tools, we light, capture, and isolate characters, weapons, and props directly from the game environment, all of which were originally created by the talented Bungie 3D art teams. Those assets are then brought into Photoshop, where they’re cut out and assembled into custom scenes using layered compositions and visual effects. The composites highlighted below are a few of the Destiny 2 pieces I’ve worked on that are meaningful to me. Let me know what your favorite is!
0:27

One of the workflows I helped advance and lead for Destiny 2 was our static compositing pipeline for marketing visuals. When we were creating marketing content for #Destiny2, we always needed strong key visuals for weapons, characters, abilities, and other in-game elements. Unlike a traditional gameplay screenshot, where everything is baked into one static image, composited shots let us isolate individual elements and light them more precisely. That gave us a lot more flexibility to build scenes with better legibility, clearer focus, stronger lighting, improved contrast, and a more intentional overall look. One of the biggest strengths of compositing is having access to individual layers in Photoshop. That layered approach not only makes the editing process easier, but also makes the final artwork much more flexible. Images can be adapted for trailer motion graphics, reworked for print, or pushed further as creative needs evolve. Destiny 2 composites are built using Adobe Photoshop alongside in-game captures. Using advanced development tools, we light, capture, and isolate characters, weapons, and props directly from the game environment, all of which were originally created by the talented Bungie 3D art teams. Those assets are then brought into Photoshop, where they’re cut out and assembled into custom scenes using layered compositions and visual effects. The composites highlighted below are a few of the Destiny 2 pieces I’ve worked on that are meaningful to me. Let me know what your favorite is!

Biwald

67,791 Aufrufe • vor 1 Monat

-- What holds it together -- ✂️PAPER CUTS.- When the image is the end ⤵️ What you bring to life is the journey towards the image: how the pieces come to be as they are. The movement is the idea, and your image in Seedance is the final frame. ---------------------------------------------------- What happens to the object? 1⃣ By the time the object TRANFORMS into something else. It takes place in a continuous shot, a single camera that zooms in slowly and never cuts away. A cut would break the transformation; the hypnotic effect lies precisely in the fact that it doesn’t flicker. Template.- [@.RE IMAGE] is the LAST frame. [GLOBAL] [Technique + light + backdrop + focus]. [Real materials]. One continuous hypnotic transformation — a single take, no cuts. The same [OBJECT A] does not appear in pieces; it [VERB: melts / folds / blooms / dissolves] into [OBJECT B]. Slow surreal dreamlike drift, one unbroken slow push, slight stop-motion shimmer, no snapping. 0-1s: [state A, intact and recognizable]. 1-2.5s: [the change begins — "the same material begins to..."]. 2.5-3.5s: [the change at full — "...becomes..."]. 3.5-5s: [it settles]. The push eases to rest. Locked, exact match to the last frame. [LOGIC RULE] one continuous same-lens push, never cutting. [A] morphs into [B], [details that must not deform] stay legible, no warping. hypnotic drift. SFX: [a sound that also transforms]. no music, normal speed. 2⃣ When one thing LEADS to another, we need each step to be a link in the chain of causality. Template.- [@.RE IMAGE] is the LAST frame. [GLOBAL] [Technique + light + backdrop + focus]. [Real materials]. Hypnotic stop-motion paper cadence, slight frame-step, brisk causal montage, ~1.5s per shot, no naturalistic motion, no slow-mo. [cut] [shot + camera] / [carries on from previous cut] / [ACTION: what it DOES, not how it looks]. SFX: [beat-anchored hit]. [cut] ... (link 2 — chains from link 1) [cut] ... (link 3) [cut] ... (link 4) [cut] ... ease back to reveal / [link 5]. Locked, exact match to the last frame. SFX: ... [LOGIC RULE].- [materials], [what must NOT happen], no warping, [text legible if any]. ~1.5s per shot, no slow-mo. no music.
0:41

-- What holds it together -- ✂️PAPER CUTS.- When the image is the end ⤵️ What you bring to life is the journey towards the image: how the pieces come to be as they are. The movement is the idea, and your image in Seedance is the final frame. ---------------------------------------------------- What happens to the object? 1⃣ By the time the object TRANFORMS into something else. It takes place in a continuous shot, a single camera that zooms in slowly and never cuts away. A cut would break the transformation; the hypnotic effect lies precisely in the fact that it doesn’t flicker. Template.- [@.RE IMAGE] is the LAST frame. [GLOBAL] [Technique + light + backdrop + focus]. [Real materials]. One continuous hypnotic transformation — a single take, no cuts. The same [OBJECT A] does not appear in pieces; it [VERB: melts / folds / blooms / dissolves] into [OBJECT B]. Slow surreal dreamlike drift, one unbroken slow push, slight stop-motion shimmer, no snapping. 0-1s: [state A, intact and recognizable]. 1-2.5s: [the change begins — "the same material begins to..."]. 2.5-3.5s: [the change at full — "...becomes..."]. 3.5-5s: [it settles]. The push eases to rest. Locked, exact match to the last frame. [LOGIC RULE] one continuous same-lens push, never cutting. [A] morphs into [B], [details that must not deform] stay legible, no warping. hypnotic drift. SFX: [a sound that also transforms]. no music, normal speed. 2⃣ When one thing LEADS to another, we need each step to be a link in the chain of causality. Template.- [@.RE IMAGE] is the LAST frame. [GLOBAL] [Technique + light + backdrop + focus]. [Real materials]. Hypnotic stop-motion paper cadence, slight frame-step, brisk causal montage, ~1.5s per shot, no naturalistic motion, no slow-mo. [cut] [shot + camera] / [carries on from previous cut] / [ACTION: what it DOES, not how it looks]. SFX: [beat-anchored hit]. [cut] ... (link 2 — chains from link 1) [cut] ... (link 3) [cut] ... (link 4) [cut] ... ease back to reveal / [link 5]. Locked, exact match to the last frame. SFX: ... [LOGIC RULE].- [materials], [what must NOT happen], no warping, [text legible if any]. ~1.5s per shot, no slow-mo. no music.

AlexandrIA

41,320 Aufrufe • vor 1 Monat

Can vision-language-action (VLA) models generalize to diverse OOD tasks and align with customized objectives? 🤔 🚀 We introduce GRAPE, a plug-and-play algorithm to generalize robot policies via preference alignment. GRAPE unfolds three benefits to boost the generalizability of VLAs: 👉1. GRAPE aligns VLAs on a trajectory level and endows the model with the ability for global decision-making, instead of merely cloning behavior; 👉2. GRAPE implicitly models reward from both successful and failed trials to boost generalizability to diverse tasks; 👉3. GRAPE adopts a scalable preference synthesis algorithm to rank trajectories with preferences that align with arbitrary objectives. Our experiments on a diverse array of real-world and simulated robotic tasks reveal: 1⃣GRAPE enhances the performance of state-of-the-art VLA models, increasing success rates on in-domain and unseen manipulation tasks by 51.79% and 60.36%; 2⃣GRAPE is versatile to be aligned with diverse objectives and reduce collision rates by 44.31% or rollout length by 11.15% when aligning towards safer or more efficient manipulation policy, respectively. Check out our full project for more details: 🔥 Paper: 🔥 Project: 🔥 Code:
1:06

Can vision-language-action (VLA) models generalize to diverse OOD tasks and align with customized objectives? 🤔 🚀 We introduce GRAPE, a plug-and-play algorithm to generalize robot policies via preference alignment. GRAPE unfolds three benefits to boost the generalizability of VLAs: 👉1. GRAPE aligns VLAs on a trajectory level and endows the model with the ability for global decision-making, instead of merely cloning behavior; 👉2. GRAPE implicitly models reward from both successful and failed trials to boost generalizability to diverse tasks; 👉3. GRAPE adopts a scalable preference synthesis algorithm to rank trajectories with preferences that align with arbitrary objectives. Our experiments on a diverse array of real-world and simulated robotic tasks reveal: 1⃣GRAPE enhances the performance of state-of-the-art VLA models, increasing success rates on in-domain and unseen manipulation tasks by 51.79% and 60.36%; 2⃣GRAPE is versatile to be aligned with diverse objectives and reduce collision rates by 44.31% or rollout length by 11.15% when aligning towards safer or more efficient manipulation policy, respectively. Check out our full project for more details: 🔥 Paper: 🔥 Project: 🔥 Code:

Huaxiu Yao

19,988 Aufrufe • vor 1 Jahr

✨A Glimpse Into The Future Of Star Atlas🤯 Check out this must-watch personal project video by Magic Puncher, Gameplay Engineer for Star Atlas, showcasing the exciting possibilities that lie ahead for Star Atlas using #UnrealEngine5. Dive into the not-too-distant future and witness the incredible potential of Star Atlas!🚀 Experience the future of the Star Atlas #Metaverse:👀 🔸Solar system scale map - Traveling seamlessly across without loading screens in #UE5 🔸600km diameter Planets 🔸500,000 Km solar system 🔸Warp Drive use traveling up to 4400km/sec 🔸Full sized planets tech 🔸Jump Gate - Starpath system with intersystem travel Watch the full video on YouTube: Want to learn more about Star Atlas? ▪️ Read the White Paper, Economic Paper, Economic Quarterly Reports, & Sustainable Governance Framework by visiting - ▪️ Join the Star Atlas Discord - ▪️ Visit the Star Atlas Marketplace - #WeAreStarAtlas #Solana #Gamefi #P2E #NFTs #Crypto #StarCitizen #Starfield #Web3gaming #BlockchainGaming
9:09

✨A Glimpse Into The Future Of Star Atlas🤯 Check out this must-watch personal project video by Magic Puncher, Gameplay Engineer for Star Atlas, showcasing the exciting possibilities that lie ahead for Star Atlas using #UnrealEngine5. Dive into the not-too-distant future and witness the incredible potential of Star Atlas!🚀 Experience the future of the Star Atlas #Metaverse:👀 🔸Solar system scale map - Traveling seamlessly across without loading screens in #UE5 🔸600km diameter Planets 🔸500,000 Km solar system 🔸Warp Drive use traveling up to 4400km/sec 🔸Full sized planets tech 🔸Jump Gate - Starpath system with intersystem travel Watch the full video on YouTube: Want to learn more about Star Atlas? ▪️ Read the White Paper, Economic Paper, Economic Quarterly Reports, & Sustainable Governance Framework by visiting - ▪️ Join the Star Atlas Discord - ▪️ Visit the Star Atlas Marketplace - #WeAreStarAtlas #Solana #Gamefi #P2E #NFTs #Crypto #StarCitizen #Starfield #Web3gaming #BlockchainGaming

Siggy

27,230 Aufrufe • vor 2 Jahren