Research map / Video / Action recognition
My View Is the Best View: Procedure Learning from Egocentric Videos
Siddhant Bansal, Chetan Arora, C.V. Jawahar
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| replace TCC with the TC3I loss (adds C-IDM temporal coherence) (ablation) | CnC · EgoProceL CMU-MMAC | F-score: 19.7 -> 22.7 | Improved |
| replace LAV with TC3I loss (ablation) | CnC · EgoProceL CMU-MMAC | F-score: 21.1 -> 22.7 | Improved |
| add TCC to LAV loss (ablation) | CnC · EgoProceL EGTEA Gaze+ | F-score: 19.1 -> 18.6 | Hurt |
| use PCM graph cut instead of hard clustering (ablation) | CnC · EgoProceL CMU-MMAC | F-score: 19.15 -> 22.72 | Improved |
| use PCM instead of subset selection (ablation) | CnC · EgoProceL CMU-MMAC | F-score: 19.66 -> 22.72 | Improved |
| increase number of key-steps K from 7 to 15 (ablation) | CnC · EgoProceL CMU-MMAC | F1: 22.7 -> 20.1 | Hurt |
| increase K from 7 to 15 for hard clustering baseline (ablation) | TC3I + HC · EgoProceL CMU-MMAC | F1: 19.2 -> 16.8 | Hurt |
| use egocentric view instead of back third-person view (ablation) | CnC · CMU-MMAC | F-score: 20.0 -> 22.7 | Improved |
| use egocentric instead of top third-person view (ablation) | CnC · CMU-MMAC | IoU: 8.1 -> 11.1 | Improved |
| CnC versus Shen et al. on ProceL with video only | CnC · ProceL | F-score: 21.1 -> 21.6 | Improved |
| CnC versus Kukleva et al. on CrossTask | CnC · CrossTask | F-score: 15.3 -> 22.6 | Improved |
| use PCM instead of hard clustering for key-step assignment | TC3I + HC · EgoProceL MECCANO | IoU: 8.0 -> 7.8 | No clear effect |
Closest papers by meaning
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.88)
- Error Detection in Egocentric Procedural Task Videos (CVPR 2024 · similarity 0.87)
- Action Scene Graphs for Long-Form Understanding of Egocentric Videos (CVPR 2024 · similarity 0.87)
- Generating 6DoF Object Manipulation Trajectories from Action Description in Egocentric Vision (CVPR 2025 · similarity 0.87)
- Retrieval-Augmented Egocentric Video Captioning (CVPR 2024 · similarity 0.87)
- Test-Time Adaptation for Combating Missing Modalities in Egocentric Videos (ICLR 2025 · similarity 0.86)
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.
