Haofei Xu

I am a PhD student at ETH Zurich and University of Tübingen, supervised by Marc Pollefeys and Andreas Geiger.

I was very fortunate to work with several amazing supervisors. I spent a wonderful time working remotely with Jianfei Cai and Hamid Rezatofighi at Monash University, Australia, prior to starting my PhD. I obtained a master's degree at University of Science and Technology of China (USTC) supervised by Juyong Zhang. During my master's, I exchanged at Nanyang Technological University (NTU), Singapore, where I was supervised by Jianfei Cai and Jianmin Zheng. I also interned at Microsoft Research Asia (MSRA), where I was mentored by Jiaolong Yang and Xin Tong.

Email  /  Scholar  /  Twitter  /  Github

photo
Selected Publications

I have broad interests in computer vision, particularly in fundamental research problems like dense correspondences, motion, 3D and video representation learning. I like to explore simple and effective approaches to solving fundamental challenges. Please see the full publication list on Google Scholar.
DepthSplat: Connecting Gaussian Splatting and Depth
Haofei Xu, Songyou Peng, Fangjinhua Wang, Hermann Blum, Daniel Barath, Andreas Geiger, Marc Pollefeys
project page / code

DepthSplat enables cross-task interactions between Gaussian splatting and depth estimation.

MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Images
Yuedong Chen, Haofei Xu, Chuanxia Zheng, Bohan Zhuang, Marc Pollefeys, Andreas Geiger, Tat-Jen Cham, Jianfei Cai
European Conference on Computer Vision (ECCV), 2024 (Oral)
project page / code

A cost volume representation for efficiently predicting 3D Gaussians from sparse multi-view images in a single forward pass.

LaRa: Efficient Large-Baseline Radiance Fields
Anpei Chen, Haofei Xu, Stefano Esposito, Siyu Tang, Andreas Geiger
European Conference on Computer Vision (ECCV), 2024
project page / code

A feed-forward 2DGS model trained in two days using four GPUs.

Unifying Flow, Stereo and Depth Estimation
Haofei Xu, Jing Zhang, Jianfei Cai, Hamid Rezatofighi, Fisher Yu, Dacheng Tao, Andreas Geiger
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2023
project page / slides / video(cn) / colab / demo / code

A unified dense correspondence matching formulation enables three motion and 3D perception tasks to be solved with a unified model.

GMFlow: Learning Optical Flow via Global Matching
Haofei Xu, Jing Zhang, Jianfei Cai, Hamid Rezatofighi, Dacheng Tao
Computer Vision and Pattern Recognition (CVPR), 2022 (Oral)
slides / video(cn) / poster / code

Learning cross-view features with a Transformer enables optical flow to be solved by directly comparing feature similarities.

High-Resolution Optical Flow from 1D Attention and Correlation
Haofei Xu, Jiaolong Yang, Jianfei Cai, Juyong Zhang, Xin Tong
International Conference on Computer Vision (ICCV), 2021 (Oral)
code

Factorizing 2D optical flow with 1D attention and 1D correlation enables 4K resolution optical flow estimation on standard GPUs.

AANet: Adaptive Aggregation Network for Efficient Stereo Matching
Haofei Xu, Juyong Zhang
Computer Vision and Pattern Recognition (CVPR), 2020
code

A sparse points-based cost aggregation method leads to an efficient and accurate stereo matching architecture without any 3D convolutions.

Invited Talks
  • DepthSplat: Connecting Gaussian Splatting and Depth, Google DeepMind, hosted by Ben Poole, 2024.10.29
  • Unifying Flow, Stereo and Depth Estimation [slides], Synced, 2022.12.28
  • GMFlow: Learning Optical Flow via Global Matching [slides], Monash University, 2022.04.13
Academic Services
  • Conference Reviewer: ICCV 2021, CVPR 2022, ECCV 2022, CVPR 2023, NeurIPS 2023, CVPR 2024, ECCV 2024, NeurIPS 2024
  • Journal Reviewer: TIP, IJCV, TPAMI
Awards

awesome website template