Welcome

Yinlin Hu (胡银林)

I am a senior researcher & engineer at Magic Leap. Previously, I was a postdoctoral scientist at EPFL-CVLab. My main focus is on developing 3D computer vision solutions for bridging the digital and physical worlds, and I am exploring effective strategies of embedding traditional geometry knowledge into modern deep learning techniques.

CV, GoogleScholar, GitHub

huyinlin@gmail.com

Image

News

  • 08.2022 I join Magic Leap as a senior researcher & engineer.
  • 10.2019 We start the research tackling vision challenges in ClearSpace-1. [Solutions]
  • 09.2019 The RicFlow developed during my PhD is merged into OpenCV. [RIC_Interpolator]
  • 01.2018 I join EPFL CVLab as a research scientist.
  • 12.2017 I pass my PhD defence.

Publications

Please find the latest list at GoogleScholar.

  • Yang Hai, Rui Song, Jiaojiao Li, David Ferstl, Yinlin Hu. Pseudo Flow Consistency for Self-Supervised 6D Object Pose Estimation. International Conference on Computer Vision (ICCV), 2023. [Code]
  • Fulin Liu, Yinlin Hu, Mathieu Salzmann. Linear-Covariance Loss for End-to-End Learning of 6D Pose Estimation. International Conference on Computer Vision (ICCV), 2023. [Code]
  • Yang Hai, Rui Song, Jiaojiao Li, Yinlin Hu. Shape-Constraint Recurrent Flow for 6D Object Pose Estimation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. [Code]
  • Shuxuan Guo, Yinlin Hu, Jose M Alvarez, Mathieu Salzmann. Knowledge Distillation for 6D Pose Estimation by Aligning Distributions of Local Predictions. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. [Code]
  • Yang Hai, Rui Song, Jiaojiao Li, Mathieu Salzmann, Yinlin Hu. Rigidity-Aware Detection for 6D Object Pose Estimation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023. [Code]
  • Yinlin Hu, Pascal Fua, Mathieu Salzmann. Perspective Flow Aggregation for Data-Limited 6D Object Pose Estimation. European Conference on Computer Vision (ECCV), 2022 [Code]
  • Chen Zhao, Yinlin Hu, Mathieu Salzmann. Fusing Local Similarities for Retrieval-based 3D Orientation Estimation of Unseen Objects. European Conference on Computer Vision (ECCV), 2022 [Code]
  • Van Nguyen Nguyen, Yinlin Hu, Yang Xiao, Mathieu Salzmann, Vincent Lepetit. Templates for 3D Object Pose Estimation Revisited: Generalization to New Objects and Robustness to Occlusions. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022. [Code]
  • Yinlin Hu, Sébastien Speierer, Wenzel Jakob, Pascal Fua, Mathieu Salzmann. Wide-Depth-Range 6D Object Pose Estimation in Space. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. [Code]
  • Zhigang Li, Yinlin Hu, Mathieu Salzmann, Xiangyang Ji. SD-Pose: Semantic Decomposition for Cross-Domain 6D Object Pose Estimation. Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), 2021.
  • Wei Wang, Zheng Dang, Yinlin Hu, Pascal Fua, Mathieu Salzmann. Robust Differentiable SVD. IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2021.
  • Zheng Dang, Kwang Moo Yi, Yinlin Hu, Fei Wang, Pascal Fua, Mathieu Salzmann. Eigendecomposition-Free Training of Deep Networks for Linear Least-Square Problems. IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2020. [Code]
  • Yinlin Hu, Pascal Fua, Wei Wang, Mathieu Salzmann. Single-Stage 6D Object Pose Estimation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020. [Code]
  • Wei Wang, Zheng Dang, Yinlin Hu, Pascal Fua, Mathieu Salzmann. Backpropagation-Friendly Eigendecomposition. Thirty-third Conference on Neural Information Processing Systems (NeurIPS), 2019. [Code]
  • Yinlin Hu, Joachim Hugonot, Pascal Fua, Mathieu Salzmann. Segmentation-driven 6D Object Pose Estimation. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019. [Code]
  • Zheng Dang, Kwang Moo Yi, Yinlin Hu, Fei Wang, Pascal Fua, Mathieu Salzmann. Eigendecomposition-Free Training of Deep Networks with Zero Eigenvalue-Based Losses. European Conference on Computer Vision (ECCV), pp. 792-807, 2018 [Code]
  • Yinlin Hu, Yunsong Li, Rui Song, Peng Rao, Yangli Wang. Minimum barrier superpixel segmentation. Image Vision Computing (IVC). Vol. 70, pp. 1-10, 2018 [Code]
  • Yunsong Li, Yinlin Hu, Rui Song, Peng Rao, Yangli Wang. Coarse-to-Fine PatchMatch for Dense Correspondence. IEEE Transactions on Circuits System and Video Technique (T-CSVT). Vol. 28, Nr. 9, pp. 2233-2245, 2018 [Code]
  • Yinlin Hu, Yunsong Li, Rui Song. Robust Interpolation of Correspondences for Large Displacement Optical Flow. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4791-4799, 2017 [Code][OpenCV version]
  • Yinlin Hu, Rui Song, Yunsong Li, Peng Rao, Yangli Wang. Highly accurate optical flow estimation on superpixel tree. Image Vision Computing (IVC), Vol. 52, pp. 167-177, 2016
  • Yinlin Hu, Rui Song, Yunsong Li. Efficient Coarse-to-Fine Patch Match for Large Displacement Optical Flow. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 5704-5712, 2016 [Code]
Search