Guanxing Lu | 卢冠行

GuanXing Lu | 卢冠行

Can be pronounced as "Kuanhsing Lu".

Hello! I am a Master's student at Tsinghua University Shenzhen International Graduate School, Tsinghua University. I am fortunate to be supervised by Prof. Yansong Tang in IVG@SZ group and Dr. Ziwei Wang from CMU. I am currently working at General Vision Lab in BIGAI as a research intern advised by Dr. Baoxiong Jia and Dr. Siyuan Huang. I received my bachelor's degree in Chien-Shiung Wu College from Southeast University, where I was a member of PALM lab and advised by Prof. Yuheng Jia and Prof. Junhui Hou.

CV / Email / Github / Zhihu / X / WeChat

"A general agent."

Selected Publications for Recent Interests [ Full List in Google Scholar ]

My research passion is on "building end-to-end general agent for navigation and manipulation" . If you would like to chat about life or research ideas on 3D vision/robotics, feel free to contact me, I am always open to these conversations.

	AnyBimanual: Transferring Unimanual Policy for General Bimanual Manipulation Guanxing Lu^, Tengbo Yu^, Haoyuan Deng, Season Si Chen, Yansong Tang^†, Ziwei Wang. Preprint 2024 [project page] / [arxiv] / [code] A simple yet effective training framework that transfers any pretrained unimanual robotic manipulation policy to multi-task bimanual manipulation policy.
	ManiCM: Real-time 3D Diffusion Policy via Consistency Model for Robotic Manipulation Guanxing Lu^, Zifeng Gao^, Tianxing Chen, Wenxun Dai, Ziwei Wang, and Yansong Tang. Preprint 2024 [project page] / [arxiv] / [code] We propose a real-time point cloud-conditioned diffusion model named ManiCM for robotic manipulation that imposes the consistency constraint to the diffusion process, so that the model can generate robot actions in only one-step inference.
	ManiGaussian: Dynamic Gaussian Splatting for Multi-task Robotic Manipulation Guanxing Lu, Shiyi Zhang, Ziwei Wang, Changliu Liu, Jiwen Lu and Yansong Tang. ECCV 2024 [project page] / [arxiv] / [code] / [中文解读] We propose a dynamic Gaussian Splatting method named ManiGaussian for multi-task robotic manipulation, which mines scene dynamics via future scene reconstruction.
	ThinkBot: Embodied Instruction Following with Thought Chain Reasoning Guanxing Lu, Ziwei Wang, Changliu Liu, Jiwen Lu and Yansong Tang. Preprint 2023 [project page] / [arxiv] In this paper, we have presented a ThinkBot agent that reasons the thought chain for missing instruction recovery in embodied instruction following (EIF) tasks.
	Semi-Supervised Subspace Clustering via Tensor Low-Rank Representation Yuheng Jia, Guanxing Lu, Hui Liu and Junhui Hou. IEEE TCSVT 2023 [arxiv] / [code] We proposed a novel semi-supervised subspace clustering method, which can simultaneously augment the initial supervisory information and construct a discriminative affinity matrix by leveraging their identical global low-rank structure.

Misc

🏊️ Swimming: I am a fan of swimming. I have learned four swimming styles by watching videos.
📖 Reading: Recently, I enjoy reading sci-fi novels. Looking for book friends in Weread!

Thanks to Jon Barron and ClustrMaps.

踏平坎坷成大道，斗罢艰险又出发。