Summary of Egooops: a Dataset For Mistake Action Detection From Egocentric Videos Referring to Procedural Texts, by Yuto Haneji et al.
EgoOops: A Dataset for Mistake Action Detection from Egocentric Videos Referring to Procedural Texts
by Yuto Haneji, Taichi Nishimura, Hirotaka Kameko, Keisuke Shirai, Tomoya Yoshida, Keiya Kajimura, Koki Yamamoto, Taiyu Cui, Tomohiro Nishimoto, Shinsuke Mori
First submitted to arxiv on: 7 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
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 A machine learning approach is proposed to detect mistakes in text-following activities, which is crucial for developing intelligent archives that provide feedback. Existing studies have focused on visually apparent mistakes, but this paper addresses the gap by introducing a new dataset called EgoOops, featuring egocentric videos and procedural texts across diverse domains. The dataset includes video-text alignment, mistake labels, and descriptions for mistakes. A mistake detection approach is also proposed, combining video-text alignment and mistake label classification to leverage the texts. Experimental results show that incorporating procedural texts is essential for accurate mistake detection. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Mistake action detection in text-following activities is important for developing intelligent archives. Most studies have focused on visually apparent mistakes, but this new approach addresses a gap by using procedural texts. The dataset features egocentric videos and texts across different areas. It includes labels that show what’s wrong with the actions. An algorithm is proposed to detect mistakes by combining video-text alignment and mistake label classification. Results show that using procedural texts makes it easier to accurately detect mistakes. |
Keywords
» Artificial intelligence » Alignment » Classification » Machine learning
