|
I am currently a PhD student at Westlake University, co-supervised by Yuliang Xiu and Gerard Pons-Moll. I am interested in modeling 3D/4D worlds from casually-captured real-world images and videos. I have previously done research internships at Tencent AI Lab, working with Xuan Wang. I completed my M.S. at Xi'an Jiaotong University and my B.S. at Xiamen University. As a Muggle dreaming of the Wizarding World, I feel so fortunate to create magic through what I am exploring on this jorney. Email | Google Scholar | GitHub | Twitter |
|
|
Equal Contribution * , Corresponding Author †, Project Lead ⚑ |
|
Yue Chen, Xingyu Chen, Anpei Chen, Gerard Pons-Moll, Yuliang Xiu arXiv, 2024 project page / arXiv / video A unified framework to probe "texture and geometry awareness" of visual foundation models. Novel view synthesis serves as an effective proxy for 3D evaluation. |
|
Yue Chen*, Xingyu Chen*⚑, Xuan Wang†, Qi Zhang, Yu Guo†, Ying Shan, Fei Wang IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023 project page / arXiv / paper / code / supplementary / video / poster A Local-to-Global robust registration strategy for bundle-adjusting NeRF, combining the global and local alignment via differentiable parameter estimation. |
|
Yue Chen*, Xuan Wang*, Xingyu Chen, Qi Zhang, Xiaoyu Li, Yu Guo†, Jue Wang, Fei Wang IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2023 project page / arXiv / paper / code / supplementary / video / poster An editable, real-time rendering, and deformable human NeRF, decomposing a dynamic human into 3D UV Volumes and a 2D appearance texture. |
|
Yue Chen, Xingyu Chen†⚑, Yicen Li IEEE International Conference on Robotics and Automation (ICRA), 2023 arXiv / code / video / poster An unsupervised approach to disentangle a scene into place, appearance, and occlusion features, handling place recognition under occlusion and varying appearance. |
|
Xingyu Chen, Qi Zhang†, Xiaoyu Li, Yue Chen, Ying Feng, Xuan Wang, Jue Wang IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022 project page / arXiv / paper / code / supplementary / video / poster A method to recover 3D scenes from tourism photos with various styles and occlusions, consistently rendering occlusion-free views under different appearance hallucinations. |
|
🪄Created Magic Making cool stuff is my jam — like rebuilding Hogwarts from Harry Potter movies or recreating lovely moments in 3D from my daily photos. 🖱️Click on the images below to render 3D, powered by Feat2GS. 
|
|
Website source code from Jon Barron
|