Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

Supervised learning has held 3D Vision back for too long. Meet RayZer — a self-supervised 3D model trained with zero 3D labels: ❌ No supervision of camera & geometry ✅ Just RGB images And the wild part? RayZer outperforms supervised methods (as 3D labels from COLMAP is noisy) 🌐... show more

Hanwen Jiang

2,379 subscribers

71,634 Aufrufe • vor 1 Jahr •via X (Twitter)

Gesundheit & Wellness Wissenschaft & Technologie Bildung

Anya Rossi• Live Now

Private livecam show

9 Kommentare

Profilbild von Hanwen Jiang

Hanwen Jiangvor 1 Jahr

🔍 How does RayZer work? It performs 3D-aware image auto-encoding, which first disentangles images into scene + camera (reconstruction), then re-entangles them back into images (rendering) and learn via RGB loss. The key is splitting the images into two sets — one set to reconstruct scene, and the other to provide supervision, which avoids trivial non-3D solutions.

Profilbild von Hanwen Jiang

Hanwen Jiangvor 1 Jahr

🤯 RayZer outperforms supervised methods — why? Turns out, 3D labels from COLMAP are noisy. GS-LRM and LVSM consistently fail on scenes of glasses, high luminance intensity, and white walls. These are cases where COLMAP usually fail. This highlights the need for self-supervised learning — and shows just how powerful it can be.

Profilbild von Hanwen Jiang

Hanwen Jiangvor 1 Jahr

RayZer is similar to video generation models philosophically: ❌ No 3D-aware architecture ❌ No 3D representation & rendering equation ❌ No 3D supervision ✅ But 3D awareness emerges. (We show more inference results)

Profilbild von Hanwen Jiang

Hanwen Jiangvor 1 Jahr

Joint work with @HaoTan5 @totoro97_ @Haian_Jin @__yuezhao__ @Sai__Bi @KaiZhang9546 @fujun_luan Kalyan Sunkavalli @qixing_huang @geopavlakos

Profilbild von Dmytro Mishkin 🇺🇦

Dmytro Mishkin 🇺🇦vor 1 Jahr

Amazing! Dare to try it in Image Matching Challenge? :)

Profilbild von Hanwen Jiang

Hanwen Jiangvor 1 Jahr

haha, I don't think it works on images with different lighting conditions now

Profilbild von relu

reluvor 1 Jahr

Super cool. I’ve been looking for pose estimation without any supervision from SfM and couldn’t find any papers! Was super surprised. I’m glad someone finally got this working

Profilbild von Jeffrey Ouyang-Zhang

Jeffrey Ouyang-Zhangvor 1 Jahr

cool work!

Profilbild von Jang Hyun (Vincent) Cho

Jang Hyun (Vincent) Chovor 1 Jahr

amazing

Ähnliche Videos

(1/N) Will this be the BERT/GPT moment for 3D vision？ Finally, unsupervised pre-training for 3D works. Led by Qitao Zhao , we present E-RayZer — a fully self-supervised 3D reconstruction model that: 🔥Matches or surpasses supervised methods like VGGT 👀Learns transferable 3D representations, outperforming CroCo, VideoMAE, and DINO 📈Scales with more unlabeled data A new recipe for scalable 3D foundation models.

(1/N) Will this be the BERT/GPT moment for 3D vision？ Finally, unsupervised pre-training for 3D works. Led by Qitao Zhao , we present E-RayZer — a fully self-supervised 3D reconstruction model that: 🔥Matches or surpasses supervised methods like VGGT 👀Learns transferable 3D representations, outperforming CroCo, VideoMAE, and DINO 📈Scales with more unlabeled data A new recipe for scalable 3D foundation models.

Hanwen Jiang

58,737 Aufrufe • vor 10 Monaten

Introducing “FlowMap”, the first self-supervised, differentiable structure-from-motion method that is competitive with conventional SfM like Colmap! IMO this solves a major missing piece for internet-scale training of 3D Deep Learning methods. 1/n

Introducing “FlowMap”, the first self-supervised, differentiable structure-from-motion method that is competitive with conventional SfM like Colmap! IMO this solves a major missing piece for internet-scale training of 3D Deep Learning methods. 1/n

Vincent Sitzmann

128,620 Aufrufe • vor 2 Jahren

Concerto Joint 2D-3D Self-Supervised Learning Emerges Spatial Representations

Concerto Joint 2D-3D Self-Supervised Learning Emerges Spatial Representations

AK

11,224 Aufrufe • vor 1 Jahr

Meet LA-Pose. Our latest model taking Wayve another step towards generalization at scale. LA-Pose employs large-scale self-supervised learning, building strong motion representations for 3D perception from 10.2 million unlabeled driving video snippets, unlike today's strongest approaches that often depend on expensive, carefully curated 3D supervision. With only a lightweight pose head and limited labelled data, LA-Pose achieves: 📷 State-of-the-art camera pose estimation 🌎 Strong zero-shot generalization across diverse driving scenarios 🏷️ Orders of magnitude less labelled data than fully supervised 3D approaches Our full blog post: Explore the full paper here:

Meet LA-Pose. Our latest model taking Wayve another step towards generalization at scale. LA-Pose employs large-scale self-supervised learning, building strong motion representations for 3D perception from 10.2 million unlabeled driving video snippets, unlike today's strongest approaches that often depend on expensive, carefully curated 3D supervision. With only a lightweight pose head and limited labelled data, LA-Pose achieves: 📷 State-of-the-art camera pose estimation 🌎 Strong zero-shot generalization across diverse driving scenarios 🏷️ Orders of magnitude less labelled data than fully supervised 3D approaches Our full blog post: Explore the full paper here:

Wayve

37,042 Aufrufe • vor 5 Monaten

🔥 #LeAVJEPA: A simple recipe for self-supervised learning from sound and vision 🔥 One shared encoder, trained without class labels. Dropping a modality makes audio & video learn a common representation. 🧵

🔥 #LeAVJEPA: A simple recipe for self-supervised learning from sound and vision 🔥 One shared encoder, trained without class labels. Dropping a modality makes audio & video learn a common representation. 🧵

Arno Solin

20,072 Aufrufe • vor 17 Tagen

HunyuanWorld-Voyager is here and fully open-source! The world’s first ultra-long-range world model with native 3D reconstruction, redefining AI-driven spatial intelligence for VR, gaming, and simulations. ✅Direct 3D Output: Exports point cloud videos to 3D formats without tools like COLMAP, enabling instant 3D application use. ✅Innovative 3D Memory: Introduces a scalable world caching mechanism, ensuring geometric consistency across any camera trajectory. ✅Top-Ranked Performance: #1 on Stanford’s WorldScore, excelling in video generation and 3D reconstruction benchmarks.( Built on HunyuanWorld 1.0, Voyager blends video generation with 3D modeling, delivering camera-controlled, high-fidelity RGB-D sequences. Control scenes via keyboard or joystick for unmatched 3D consistency. Explore now: 🌐Project Page: 🔗GitHub: 🤗HuggingFace: 📝Technical Details:

HunyuanWorld-Voyager is here and fully open-source! The world’s first ultra-long-range world model with native 3D reconstruction, redefining AI-driven spatial intelligence for VR, gaming, and simulations. ✅Direct 3D Output: Exports point cloud videos to 3D formats without tools like COLMAP, enabling instant 3D application use. ✅Innovative 3D Memory: Introduces a scalable world caching mechanism, ensuring geometric consistency across any camera trajectory. ✅Top-Ranked Performance: #1 on Stanford’s WorldScore, excelling in video generation and 3D reconstruction benchmarks.( Built on HunyuanWorld 1.0, Voyager blends video generation with 3D modeling, delivering camera-controlled, high-fidelity RGB-D sequences. Control scenes via keyboard or joystick for unmatched 3D consistency. Explore now: 🌐Project Page: 🔗GitHub: 🤗HuggingFace: 📝Technical Details:

Tencent Hy

198,391 Aufrufe • vor 1 Jahr

Introducing DINOv3: a state-of-the-art computer vision model trained with self-supervised learning (SSL) that produces powerful, high-resolution image features. For the first time, a single frozen vision backbone outperforms specialized solutions on multiple long-standing dense prediction tasks. Learn more about DINOv3 here:

Introducing DINOv3: a state-of-the-art computer vision model trained with self-supervised learning (SSL) that produces powerful, high-resolution image features. For the first time, a single frozen vision backbone outperforms specialized solutions on multiple long-standing dense prediction tasks. Learn more about DINOv3 here:

AI at Meta

901,780 Aufrufe • vor 1 Jahr

Spatio-Temporal Reconstruction Model for Large-Scale Outdoor Scenes Contributions: • We propose STORM, the first feed-forward, self-supervised method for fast and accurate reconstruction of dynamic 3D scenes from sparse, multi-timestep, posed camera images. • Our bottom-up framework aggregates and transforms per-frame 3D Gaussian Splats into a cohesive scene representation, enabling self-supervised motion estimation. Furthermore, we introduce motion tokens that capture common motion primitives and regularize motion predictions, facilitating dynamic motion group segmentation without explicit motion or correspondence supervision. • We present several enhancements for in-the-wild scenarios, including sky modeling, camera exposure inconsistency handling, large novel-view extrapolation, and fine-grained human motions reconstruction, making STORM well-suited for real-world applications.

Spatio-Temporal Reconstruction Model for Large-Scale Outdoor Scenes Contributions: • We propose STORM, the first feed-forward, self-supervised method for fast and accurate reconstruction of dynamic 3D scenes from sparse, multi-timestep, posed camera images. • Our bottom-up framework aggregates and transforms per-frame 3D Gaussian Splats into a cohesive scene representation, enabling self-supervised motion estimation. Furthermore, we introduce motion tokens that capture common motion primitives and regularize motion predictions, facilitating dynamic motion group segmentation without explicit motion or correspondence supervision. • We present several enhancements for in-the-wild scenarios, including sky modeling, camera exposure inconsistency handling, large novel-view extrapolation, and fine-grained human motions reconstruction, making STORM well-suited for real-world applications.

MrNeRF

53,292 Aufrufe • vor 1 Jahr

Introducing Neural Jacobian Fields, robot 3D kinematic models learned only from vision! They can model & control robots from just a single RGB camera, even those w/ intractable kinematics & no embedded sensors such as soft, 3D-printed pneumatic hands! 1/n

Introducing Neural Jacobian Fields, robot 3D kinematic models learned only from vision! They can model & control robots from just a single RGB camera, even those w/ intractable kinematics & no embedded sensors such as soft, 3D-printed pneumatic hands! 1/n

Vincent Sitzmann

54,285 Aufrufe • vor 2 Jahren

Introducing FLARE #CVPR2026 2025 FLARE is a feed-forward model that simultaneously estimates high-quality camera poses, 3D geometry, and appearance from sparse uncalibrated images. 1/4

Introducing FLARE #CVPR2026 2025 FLARE is a feed-forward model that simultaneously estimates high-quality camera poses, 3D geometry, and appearance from sparse uncalibrated images. 1/4

Gordon Wetzstein

29,572 Aufrufe • vor 1 Jahr

We present VLM-3R: a Vision-Language Model capable of 3D spatial reasoning from monocular video, grounding visual cues, geometry, and camera motion. ✅ No depth sensor ✅ No pre-built 3D maps ✅ End-to-end spatial + temporal reasoning 🔗 Code & benchmark: #VLM #3DVision #LLMs

We present VLM-3R: a Vision-Language Model capable of 3D spatial reasoning from monocular video, grounding visual cues, geometry, and camera motion. ✅ No depth sensor ✅ No pre-built 3D maps ✅ End-to-end spatial + temporal reasoning 🔗 Code & benchmark: #VLM #3DVision #LLMs

Zhiwen(Aaron) Fan

14,895 Aufrufe • vor 1 Jahr

[1/N] Current visual geometry prediction models primarily rely on labeled 3D data. Our CVPR26 paper, Flow3r, allows additionally leveraging unlabeled videos (using flow supervision) for scalable visual geometry learning, enabling accurate multi-view 3D reconstruction in-the-wild.

[1/N] Current visual geometry prediction models primarily rely on labeled 3D data. Our CVPR26 paper, Flow3r, allows additionally leveraging unlabeled videos (using flow supervision) for scalable visual geometry learning, enabling accurate multi-view 3D reconstruction in-the-wild.

Shubham Tulsiani

17,487 Aufrufe • vor 8 Monaten

Apple drops a new 3D model generator, LiTo > Learns geometry + appearance together in a unified 3D latent space > Turns RGB-D images into a compact surface light field > Recreates specular highlights under complex lighting > Generates 3D objects from a single image with input-consistent materials Big step toward more realistic 3D generation

Apple drops a new 3D model generator, LiTo > Learns geometry + appearance together in a unified 3D latent space > Turns RGB-D images into a compact surface light field > Recreates specular highlights under complex lighting > Generates 3D objects from a single image with input-consistent materials Big step toward more realistic 3D generation

AI Search

13,078 Aufrufe • vor 5 Monaten

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

Alexandre Morgand

15,192 Aufrufe • vor 8 Monaten

Starting the new year without human labeling 🎉!! Multimodal lidar-camera data is a gold mine of dense 3D geometry hiding in plain sight. For supervised pretraining and validation at scale at Torc-Robotics, we rely on fully automated pseudo-labeling pipelines. Exploiting geometric priors from temporally accumulated LiDAR maps and an iterative update rule enforces joint geometric–semantic consistency while detecting moving objects via inconsistencies. We achieve 3D semantic labels and 3D bounding boxes with human-like quality at 200m+ range required for highway driving. Paper: Exciting work with Torc-Robotics with Filippo Ghilotti, Samuel Brucker, Nahku Saidy, Matteo Matteucci, Mario Bijelic.

Starting the new year without human labeling 🎉!! Multimodal lidar-camera data is a gold mine of dense 3D geometry hiding in plain sight. For supervised pretraining and validation at scale at Torc-Robotics, we rely on fully automated pseudo-labeling pipelines. Exploiting geometric priors from temporally accumulated LiDAR maps and an iterative update rule enforces joint geometric–semantic consistency while detecting moving objects via inconsistencies. We achieve 3D semantic labels and 3D bounding boxes with human-like quality at 200m+ range required for highway driving. Paper: Exciting work with Torc-Robotics with Filippo Ghilotti, Samuel Brucker, Nahku Saidy, Matteo Matteucci, Mario Bijelic.

Felix Heide

18,250 Aufrufe • vor 9 Monaten

3D-VLA A 3D Vision-Language-Action Generative World Model Recent vision-language-action (VLA) models rely on 2D inputs, lacking integration with the broader realm of the 3D physical world. Furthermore, they perform action prediction by learning a direct mapping from

3D-VLA A 3D Vision-Language-Action Generative World Model Recent vision-language-action (VLA) models rely on 2D inputs, lacking integration with the broader realm of the 3D physical world. Furthermore, they perform action prediction by learning a direct mapping from

AK

85,570 Aufrufe • vor 2 Jahren

What if we can simulate an *interactive 3D world*, from a single image, in the wild, in real time? Introducing PointWorld-1B: a large pre-trained 3D world model that predicts env dynamics given RGB-D capture and robot actions. 🌐 from Stanford University NVIDIA

What if we can simulate an interactive 3D world, from a single image, in the wild, in real time? Introducing PointWorld-1B: a large pre-trained 3D world model that predicts env dynamics given RGB-D capture and robot actions. 🌐 from Stanford University NVIDIA

Wenlong Huang

279,427 Aufrufe • vor 9 Monaten

Wow. No more waiting hours for COLMAP / SfM — you get live 3D Gaussian Splats straight from unposed images, even for massive scenes. Your 3D scene and camera poses are ready right as you finish capturing. Inria team cooked again; code drops end of May.

Wow. No more waiting hours for COLMAP / SfM — you get live 3D Gaussian Splats straight from unposed images, even for massive scenes. Your 3D scene and camera poses are ready right as you finish capturing. Inria team cooked again; code drops end of May.

Bilawal Sidhu

48,530 Aufrufe • vor 1 Jahr