Yang You

🎯 Join us at #ICCV2025! The 1st ICCV Workshop & Challenge on Category-Level Object Pose Estimation in the Wild (WCLOP 2025) is happening soon! 📅 Oct 20 📍 Room 324

🚀 Excited to announce that DOT-Sim is accepted to #ICRA2026! We build a differentiable optical tactile simulator that captures both the physics and optics of soft sensors: an MPM-based elastic model handles large, non-linear deformations, and a learned residual renderer maps depth+normal to realistic sensor images. It calibrates to a real DenseTact in minutes, and enables zero-shot sim-to-real on object classification (85%), embedded tumor detection (90%), and sub-mm trajectory following. #ICRA2026 #Robotics #TactileSensing #Simulation #Sim2Real #AI Joint work with amazing collaborators: Won Kyung Do, Aiden Swann, Rika Antonova, Monroe Kennedy, and Leonidas Guibas. 🔗 ArXiv: arxiv.org/abs/2604.27367

Generative models can create visually stunning 3D rooms, but are they functional for the agents inside them? 🛋️🤖 Introducing SceneTeract - a framework that verifies 3D scene functionality under agent-specific constraints! 📄: arxiv.org/abs/2603.29798 🌐: sceneteract.github.io

Spatial representations are central to world models🌍 SuperDec is an extremely compact 3D scene representation (replacing millions of Gaussians with just a few hundred primitives) ideal for abstract reasoning and planning in 3D ➡️super-dec.github.io ✨Oral @ICCVConference

Excited to share that we’ll be presenting three works at #CoRL 2025 (@corl_conf), taking place September 27–30 in Seoul, Korea! 🎉 1. SIREN: siren-robot.github.io 2. ParticleFormer: suninghuang19.github.io/particleformer… 3. ARCH: long-horizon-assembly.github.io

🚀 Excited to announce that Img2CAD is accepted to #SIGGRAPHAsia2025! We reverse engineer 3D CAD from a single 2D image by factorizing the task: a finetuned VLM predicts discrete CAD programs, and a TrAssembler regresses continuous parameters. It works well on ShapeNet common objects like chairs and tables, and generalizes to images in the wild. #SIGGRAPHAsia2025 #3D #CAD #ComputerGraphics #VisionLanguage #AI 🔗 ArXiv: arxiv.org/abs/2408.01437 🖥️ Project: qq456cvb.github.io/projects/img2c… 💻 Code: github.com/qq456cvb/Img2C…

Call for Papers & Challenge Participants | WCLOP @ ICCV 2025 📆 Paper Deadline: 22 September 2025 (AoE) 🔔 Notification: 29 September 2025 📍 Venue: Honolulu, Hawaii We’re hosting the cross-disciplinary workshop “WCLOP: 1st ICCV Workshop and Challenge on Category-Level Object Pose Estimation in the Wild” at ICCV 2025, spotlighting advances in category-level object pose estimation for robotic manipulation scenarios. We warmly invite non-archival submissions: • Short Papers — up to 4 pages • Regular Papers — up to 8 pages Join the WCLOP 2025 Challenge as well: Two-phase evaluation on Omni6DPose & PACE datasets, followed by downstream robot-manipulation tasks 📆 Phase 1 deadline: 15 September 2025 💰 Over $6,000 in awards for challenge winners Several leading experts will deliver keynote talks. Let’s push the boundaries of category-level pose estimation 🤝 🔗 Website: wclop.github.io 🔗 Paper Submission: openreview.net/group?id=thecv… 🔗 Challenge Portal: codabench.org/competitions/9…

We made a @gradio demo for AllTracker! AllTracker is the current state-of-the-art for general-purpose point tracking. The demo gives a good sense of the accuracy---try your own videos and see for yourself! 🔗 Demo: huggingface.co/spaces/aharley… 💻 Code: github.com/aharley/alltra…

AllTracker: Efficient Dense Point Tracking at High Resolution If you're using any point tracker in any project, this is likely a drop-in upgrade—improving speed, accuracy, and density, all at once.

Excited to share our work: Gaussian Mixture Flow Matching Models (GMFlow) github.com/lakonik/gmflow GMFlow generalizes diffusion models by predicting Gaussian mixture denoising distributions, enabling precise few-step sampling and high-quality generation.

🚀 Excited to announce that our work on enhancing the 3D awareness of Vision Transformers (ViT) is accepted to #ICLR2025! We evaluate ViT models' 3D equivariance and introduce a simple finetuning strategy using 3D correspondences, achieving remarkable improvements in pose estimation, tracking, and semantic transfer even with minimal finetuning. 🔗 Arxiv paper: arxiv.org/abs/2411.19458 💻 Code is now available on github.com/qq456cvb/3DCor…, check it out! 🤗 Try out the demo on Hugging Face: huggingface.co/spaces/qq456cv… 🙏 Grateful to my amazing co-authors Yixin Li, Congyue Deng (@CongyueD), Yue Wang (@yuewang314), Leonidas Guibas for the contributions! #ICLR2025 #3Dvision

🚀 Excited to share our PACE dataset at #ECCV2024! 📊 PACE (Pose Annotations in Cluttered Environments) is a large-scale benchmark for 3D pose estimation in real-world, cluttered scenes with 55K frames, 258K annotations, and 238 objects across 43 categories. It includes PACE-Sim with 100K simulated frames and 2.4M annotations. We evaluate state-of-the-art methods in both pose estimation and tracking, revealing their limits in real-world scenarios. Also, an innovative 3-camera annotation system boosts pose labeling efficiency. 🗓️ Poster #191 on Fri, Oct 4, 10:30am-12:30pm CEST at the Exhibition Area! 🔗 Dataset: huggingface.co/datasets/qq456… 🌐 Project: qq456cvb.github.io/projects/pace 💻 Code: github.com/qq456cvb/PACE 🎥 Video: youtu.be/RX1K-xA99ZI #3Dvision #PoseEstimation #ECCV2024 #eccvconf

Siggraph is fun!#SIGGRAPH2024

Excited to share our work RPMArt, accepted by #IROS2024 🎉🎉🎉! RPMArt takes a single-view noisy point cloud as input and is able to estimate the articulation parameters and manipulate the articulation part robustly by voting. 👉Explore r-pmart.github.io for more! #Robotics

To mitigate the influence of erroneous tuple samples, we introduce a noisy pair filtering module, enabling the orientation vote to be cast exclusively by reliable point tuples.

Check out our work CPPF++ that is accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence (#TPAMI)! Are you seeking a robust, generalized 6D pose estimator for RGB-D images captured by any sensor, tailored to your custom setup? Here it is!