Summary of Solution For Ood-cv Workshop Ssb Challenge 2024 (open-set Recognition Track), by Mingxu Feng et al.
Solution for OOD-CV Workshop SSB Challenge 2024 (Open-Set Recognition Track)
by Mingxu Feng, Dian Chao, Peng Zheng, Yang Yang
First submitted to arxiv on: 30 Sep 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed hybrid approach in this report combines post-hoc OOD detection techniques with Test-Time Augmentation strategies to achieve strong performance on open-set recognition tasks. The method is evaluated using the Semantic Shift Benchmark and achieves a balance between AUROC and FPR95 scores, ranking 2nd overall in the OSR Challenge at the OOD-CV Workshop during ECCV 2024. Key components include fusion of various post-hoc techniques, different Test-Time Augmentation strategies, and evaluation of base models’ impact on final performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about finding ways to recognize things that are new or not like what a computer was trained to see before. It’s like showing a child a picture they’ve never seen before and asking if it’s a cat or dog. The authors tried different methods to do this, combining them in a special way to get the best results. They used pictures from two big datasets, ImageNet1k and ImageNet21k, to test their approach. Their method worked well, getting high scores on a benchmark called Semantic Shift Benchmark. |
