Research map / Video / Action recognition
COIN: A Large-Scale Dataset for Comprehensive Instructional Video Analysis
Check 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 task-consistency refinement to SSN-Fusion | SSN on COIN · COIN step localization, IoU 0.5 | mAP: 8.12 -> 9.05 | Helped |
| Add task-consistency refinement to SSN-Fusion | SSN on COIN · COIN step localization, IoU 0.5 | mAR: 26.79 -> 29.79 | Helped |
| Add task-consistency refinement | R-C3D on COIN · COIN step localization, IoU 0.5 | mAP: 2.82 -> 3.08 | Helped |
| Add task-consistency refinement to SSN | SSN on Breakfast · Breakfast step localization, IoU 0.5 | mAP: 15.84 -> 16.39 | Helped |
| Add task-consistency refinement to SSN | SSN on Breakfast · Breakfast step localization, IoU 0.5 | mAR: 35.51 -> 36.20 | Helped |
| Use ordering information with weak supervision | Action segmentation on COIN · COIN action segmentation | frame accuracy: 4.94 -> 21.17 | Helped |
| Use ordering information with weak supervision | Action segmentation on COIN · COIN action segmentation | frame accuracy: 4.94 -> 34.30 | Helped |
| Use frame-level labels for fully supervised classification instead of Action-Sets supervision | CNN on COIN · COIN action segmentation | frame accuracy: 4.94 -> 25.79 | Helped |
| Include temporal information and ordering constraints rather than using frame labels alone | Action segmentation on COIN · COIN action segmentation | frame accuracy: 25.79 -> 34.30 | Helped |
| Use task-consistency refinement | SSN+TC-Fusion · COIN step localization, sports domain, IoU 0.1 | mAP | Helped |
Closest papers by meaning
- Collaborative Weakly Supervised Video Correlation Learning for Procedure-Aware Instructional Video Analysis (AAAI 2024 · similarity 0.86)
- Unsupervised Learning From Narrated Instruction Videos (CVPR 2016 · similarity 0.86)
- Watch and Learn: Learning to Use Computers from Online Videos (CVPR 2026 · similarity 0.84)
- From Raw Video to Pedagogical Insights: A Unified Framework for Student Behavior Analysis (AAAI 2024 · similarity 0.83)
- Step Differences in Instructional Video (CVPR 2024 · similarity 0.82)
- SLVMBench: Skill Learning from Video Memory (NeurIPS 2026 · similarity 0.82)
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.