Deep learning models are usually developed and tested under the implicit assumption that the training and test data are drawn independently and identically distributed (IID) from the same distribution. Overlooking out-of-distribution (OOD) images can result in poor performance in unseen or adverse viewing conditions, which is common in real-world scenarios.

In this workshop, we are interested in discussing the performance of computer vision models on OOD images which follows a different distribution than the training images. Also given the current trend of web-scale pretrained computer vision models, it is of interest to better understand their performance in the OOD or rare scenarios.

Our workshop will be featuring three competitions, OOD generalization on the OOD-CV dataset, open-set recognition and generalized category discovery on the Semantic-shift benchmark.

Submission Information

We invite submissions of both long and short papers on the topic of out-of-distribution generalization in computer vision. Long papers are limited to 14 pages, and the submission deadline is August 10th, 2024 (AoE). Short papers are limited to 4 pages, and the submission deadline is August 25th, 2024 (AoE). Long papers should use the ECCV template. Short papers should use the CVPR template. Only accepted long papers will be included in the ECCV 2024 proceedings. Both accepted long and short papers will be presented as either an oral or poster presentation. At least one author of each accepted submission must present the paper at the workshop. The topics include but are not limited to:
  • Discussion of OOD generalization in the context of internet scale pretrained models
  • Improving generalization of computer vision systems in OOD scenarios
  • Research at the intersection of biological and machine vision
  • Generative causal models for image analysis
  • Domain generalization
  • Novel architectures with robustness to occlusion, viewpoint and other real-world domain shifts
  • Domain adaptation techniques for robust vision system in the real world
  • Datasets for evaluating model robustness
Please see here for submission informations.

Workshop Program

Time and location: 09/30 AM, @room Brown 1 Zoom meeting room: https://ed-ac-uk.zoom.us/j/88382108647 Password: c82HgVmf
9:00 Opening
9:10 Invited talk: Sara Beery 30mins+10mins Q&A
9:50 Invited talk: Jiahui Liu 30mins+10mins Q&A
10:30 Coffee Break
11:00 Invited talk: Zsolt Kira 30mins+10mins Q&A
11:40 Oral Presentation #1 Robust Fine-tuning and Adaptation of Zero-shot Models via Adaptive Weight-space Ensembling
11:50 Oral Presentation #2 Improving Generalization in Visual Reasoning via Self-Ensemble
12:00 Oral Presentation #3 BelHouse3D: A Benchmark Dataset for Assessing Occlusion Robustness in 3D Point Cloud Semantic Segmentation
12:10 Challenge Introduction 5mins
12:20 Challenge Winners Presentation 5mins each team
12:30 Poster Session

Invited Speakers

Organizing Committee

Program Committee

Kibok Lee, Yulong Cao, Shuo Chen, Lifeng Huang, Haoran Wang, Yuxiang Lai, Angtian Wang, Salah Ghamizi, Xin Wen, Xiaoding Yuan, Pengliang Ji, Zexin He, Umar Khalid, Jiahui Liu, Guofeng Zhang, Zihao Xiao, Jike Zhong, Junfei Xiao, Alexander Robey, Wei Hao, Junbo Li, Ziyun Li
Please contact ood-cv-challenge@googlegroups.com or zhaobc.gm@gmail.com if you have any questions.