Weize Li

Calling for Competing Teams! Help push the frontier of Bimanuals Robotics at the 2nd What Bimanuals Can Do (WBCD) at ICRA2026! - 3 concurrent locations (Vienna, Shanghai, Bay Area) - 3 real-world applications - Simulation & UMI Support - $200k Prize Pool A thread 🧵

LiteVGGT: Boosting Vanilla VGGT via Geometry-aware Cached Token Merging Zhijian Shu, @_cheng_lin, Tao Xie, Wei Yin, Ben Li, Zhiyuan Pu, @WeizeLi24, Yao Yao, Xun Cao, @gingertata, @xxlong0 tl;dr: pixel gradient+token variance->geometric importance->geometry-aware feature map->partition & group & merge tokens arxiv.org/abs/2512.04939

Come check out our poster booth on the morning of the 4th and say hi! Fri Oct 4 | 10.30 to 12.30 | Poster Session #7 | Poster Board ID #28 by the way, I'm focusing on 3DV + Robotics for Embodied AI...maybe we can talk about how to get into the new field~! #ECCV2024

Introducing our team members at #AnySyn3D! Explore our recent works with our collaborative partners. If you’re interested in connecting or partnering with us, please don’t hesitate to reach out via email at anysyn3d@gmail.com. website: anysyn3d.github.io #AIGC #AI #3D

Excited to share our #CVPR2024 highlight paper "Move as You Say, Interact as You Can", which employs scene affordance as an intermediate representation for language-guided human motion generation. Project: afford-motion.github.io Paper: arxiv.org/abs/2403.18036

I wrote a tutorial on diffusion models for undergrad and grad students. I tried my best to give intuitive explanations for complicated equations. Your feedback is much appreciated Thanks to those who suggested various reading materials to me arxiv.org/abs/2403.18103

GaussianGrasper: 3D Language Gaussian Splatting for Open-vocabulary Robotic Grasping. Checkout our page mrsecant.github.io/GaussianGraspe…… for more details. Compared to LERF, faster, more accurate 3D seg, more robust grasp. #GaussianSplatting #robotics #grasp

Introducing 𝐌𝐨𝐛𝐢𝐥𝐞 𝐀𝐋𝐎𝐇𝐀🏄 -- Hardware! A low-cost, open-source, mobile manipulator. One of the most high-effort projects in my past 5yrs! Not possible without co-lead @zipengfu and @chelseabfinn. At the end, what's better than cooking yourself a meal with the 🤖🧑‍🍳

Check out our poster on Pose-agnostic Anomaly Detection (PAD) @NeurIPSConf and see how we leverage #NeRF to detect🔍anomalies on production lines. No need for point clouds or scanners, just a single camera📸 for comprehensive 360-degree object inspection.

Multimodal reasoning is hard. Even the best LMMs struggle with counting😥 Any fix for it? Introduce VPD from @GoogleAI: we teach LMMs multimodal CoT reasoning with data synthesized from LLM + vision tools, and achieve new SOTAs on many multimodal tasks!🥳 arxiv.org/abs/2312.03052

Check out PAD! A fresh dataset & benchmark for pose-agnostic anomaly detection in object recognition. Comes with Multi-pose Anomaly Detection dataset & OmniposeAD method. Dive in for more! #AI #llm #arXiv arxiv.org/abs/2310.07716…"

Such a cool paper! Jointly optimizing shape parameters, albedo, and roughness from a few reference images through differentiable signed distance function rendering. By @DelioVicini, Sébastien Speierer, and @wenzeljakob. rgl.epfl.ch/publications/V…

I wanted to model my hotel room this week and decided to give @Polycam3D's roomplan mode a whirl. 👏 Well done! They added a wonderful touch by putting 3D objects in the output that are actual representations of the room objects. Not just geometric shapes. Why aren't more companies using Roomplan in their apps? #Computervision #AEC

We are adding support for plugins to ChatGPT — extensions which integrate it with third-party services or allow it to access up-to-date information. We’re starting small to study real-world use, impact, and safety and alignment challenges: openai.com/blog/chatgpt-p…