Research map / Video / Action recognition
Unsupervised Co-Activity Detection from Multiple Videos Using Absorbing Markov Chain
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add intra-sequence edges (temporal locality) (ablation) | AMC · YouTube co-activity, all videos | F-measure: .73 -> .78 | Helped |
| add absorbing state (vs PageRank) (ablation) | AMC · YouTube co-activity, all videos | F-measure: .49 -> .78 | Helped |
| add intra-sequence edges (ablation) | AMC · YouTube co-activity, video pairs | F-measure: .64 -> .71 | Helped |
| add absorbing state (vs PageRank) (ablation) | AMC · YouTube co-activity, video pairs | F-measure: .36 -> .71 | Helped |
| add intra-sequence edges (ablation) | AMC · THUMOS14, all videos | F-measure: .39 -> .40 | Helped |
| add absorbing state (vs PageRank) (ablation) | AMC · THUMOS14, all videos | F-measure: .19 -> .40 | Helped |
| AMC instead of CS | co-activity detection · YouTube co-activity, all videos | F-measure: .43 -> .78 | Helped |
| AMC instead of CS | co-activity detection · THUMOS14, video pairs | F-measure: .32 -> .35 | Mixed |
| increase subsequence length from 30 to 60/90 (ablation) | AMC · YouTube co-activity, all videos | F-measure: .78 -> .77 (60), .75 (90) | No clear effect |
| reduce subsequence overlap (subsampling 10 to 30) (ablation) | AMC · YouTube co-activity, all videos | F-measure: .78 -> .77 | No clear effect |
Closest papers by meaning
- Unsupervised Action Segmentation by Joint Representation Learning and Online Clustering (CVPR 2022 · similarity 0.87)
- Fast Temporal Activity Proposals for Efficient Detection of Human Actions in Untrimmed Videos (CVPR 2016 · similarity 0.87)
- METAL: Minimum Effort Temporal Activity Localization in Untrimmed Videos (CVPR 2020 · similarity 0.86)
- Learning Latent Super-Events to Detect Multiple Activities in Videos (CVPR 2018 · similarity 0.86)
- Learning to Predict Activity Progress by Self-Supervised Video Alignment (CVPR 2024 · similarity 0.84)
- Unsupervised Action Segmentation via Fast Learning of Semantically Consistent Actoms (AAAI 2024 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.