Reading Group in Computer and Robot Vision
- The purpose of this reading group is to provide an overview of the rapidly evolving computer vision literature. We meet once a week to discuss an article that all participants should read before the seminar.
- All participants are welcome to suggest, and present articles of their choosing. The theme for articles should be computer and robot vision, with emphasis on new, high-impact conference papers (e.g. from ICCV, CVPR, RSS, or ECCV). To make best use of time, you may consider choosing an article that relates to your work, that you presumably would read anyway. If the article is well written that is also a plus.
You have the option to attend the seminars as part of a PhD course. You will get one 1hp for each time you present a paper, and participate in another three seminars. If you want to go for the PhD course option, let me know in advance by sending me an email with your personal number, so I can register your attendance.
Meeting room: Visionen, Stora Konf.rummet, Campus Valla, Building B.
Time: Thursdays at 14.00-15.00.
E-mail list: Upcoming meetings and articles are announced on the mailing list vision-seminars.
- The article club starts again in January. In the meantime, you are welcome to submit suggestions of new papers to read!
- Have a look at e.g. the following proceedings: ICCV'21, CVPR'21, ECCV'20, NeurIPS'20, SIGGRAPH'21, RSS'2021, ACCV'21. Some old, unused suggestions are listed below.
- Oznan Sener and Vladlen Koltun, "Multi-task Learning as Multi-Objective Optimization", NeurIPS18 [PDF] SD201106
- Qianqian Wang et al., "Learning Feature Descriptors using Camera Pose Supervision", ECCV20 [PDF] [GIT] SD200812
- Ruiqu Gao et al. "Flow Contrastive Estimation of Energy-Based Models", CVPR20 [PDF] SD200630
- A. Ilyas et al., "Adversarial Examples Are Not Bugs, They Are Features", ArXiv 2019 [PDF] [blog post] SD190515
- M. Nakada et al., Deep Learning of Biomimetic Sensorimotor Control for Biomechanical Human Animation, TOG 2018 [PDF] SD201218
- T. Takikawa et al., Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Surfaces ArXiv'21 [PDF] [GIT] SD210130
- A. Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale, [ICLR21] SD210224
- M. Charon et al., Emerging Properties in Self-Supervised Vision Transformers, ArXiv'21 [PDF] [blog] [GIT] SD210506
Paper log spring 2021
- Jan 27: Oliver presents: W. Hamilton et al. Inductive Representation Learning on Large Graphs, NeurIPS'17 [PDF][Git][project page]
- Feb 10: Johan presents: Z. Allen-Zhu and Y. Li, Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning, ArXiv'20 preprint [PDF] [blog post]
- Feb 24: Arvi presents: Y. Wang et al., Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation, CVPR2020 [PDF] [GIT]
- March 24: Johan presents: H. Pham et al. Meta Pseudo Labels [ArXiv] [GIT]
- March 31: Joakim presents: A. Jaegle et al., Perceiver: General Perception with Iterative Attention, ArXiv 2021, [PDF]
- April 7: Andreas presents: S. Wizadwongsa et al., NeX: Real-time View Synthesis with Neural Basis Expansion CVPR21, [GIT]
- April 21: Johan presents: S. Gould et al. Deep Declarative Networks, IEEE TPAMI [PDF]
- May 5: Joakim presents: C. Yang et al. Self-supervised Video Object Segmentation by Motion Grouping, [ArXiv] [project page]
- May 12: Pavlo presents: M. Finzi et al. A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups, ArXiv'21 [PDF] [GIT]
Paper log autumn 2021
- Oct 7: Per-Erik presents: O. Limoyo et al. Heteroscedastic Uncertainty for Robust Generative Latent Dynamics, RAL July 2020, [PDF] [GIT]
- Oct 14: Joakim presents: T. Meinhardt et al., TrackFormer: Multi-Object Tracking with Transformers, ArXiV'21 [PDF] [GIT]
- Oct 28: Mårten presents: Y. Ding et al., Minimal Solutions for Panoramic Stitching Given Gravity Prior, ICCV 2021 [PDF]
- Nov 4, Johan presents: M. Cisse et al., mixup: Beyond Empirical Risk Minimization, ICLR2018, [PDF] [reviews]
- Dec 9, Emil presents: Y. Zhao et al., Deep Lucas-Kanade Homography for Multimodal Image Alignment, CVPR2021, [PDF] [Git]
Old paper logs
Senast uppdaterad: 2021-12-12