Research map / Video / Action recognition
Interactive Prototype Learning for Egocentric Action Recognition
Xiaohan Wang, Linchao Zhu, Heng Wang, Yi Yang
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 |
|---|---|---|---|
| add IPL interaction instead of NetVLAD (ablation) | R(2+1)D-34 · EPIC-KITCHENS-100 val | action top-1 accuracy: 38.80 -> 40.98 | Improved |
| add IPL interaction instead of NetVLAD (ablation) | R(2+1)D-34 · EPIC-KITCHENS-100 val | verb top-1 accuracy: 67.39 -> 68.61 | Improved |
| add verb-to-noun top-K selection (ablation) | IPL R(2+1)D · EPIC-KITCHENS-100 val | noun top-1 accuracy: 49.68 -> 51.24 | Improved |
| add verb-to-noun top-K selection (ablation) | IPL R(2+1)D · EPIC-KITCHENS-100 val | verb top-1 accuracy: 66.82 -> 68.61 | Improved |
| add background prototype center (ablation) | IPL R(2+1)D · EPIC-KITCHENS-100 val | noun top-1 accuracy: 50.30 -> 51.24 | Improved |
| add IPL framework | I3D · EPIC-KITCHENS-100 val | noun top-1 accuracy: 48.48 -> 50.87 | Improved |
| add IPL framework | R(2+1)D-34 · EPIC-KITCHENS-100 val | noun top-1 accuracy: 47.55 -> 51.24 | Improved |
| add IPL framework | R(2+1)D-34 · EPIC-KITCHENS-100 val | action top-1 accuracy: 37.62 -> 40.98 | Improved |
| add IPL framework | I3D · EPIC-KITCHENS-55 val | noun top-1 accuracy: 31.3 -> 33.2 | Improved |
| add IPL framework | R(2+1)D-34 · EPIC-KITCHENS-55 val | noun top-1 accuracy: 31.1 -> 35.5 | Improved |
| add IPL framework | I3D · EGTEA (avg of 3 splits) | mean class accuracy: 55.21 -> 59.05 | Improved |
Closest papers by meaning
- Opening the Vocabulary of Egocentric Actions (NeurIPS 2023 · similarity 0.90)
- Learning to Recognize Actions on Objects in Egocentric Video With Attention Dictionaries (TPAMI 2021 · similarity 0.90)
- Ego-Only: Egocentric Action Detection without Exocentric Transferring (ICCV 2023 · similarity 0.90)
- Symbiotic Attention for Egocentric Action Recognition With Object-Centric Alignment (TPAMI 2020 · similarity 0.89)
- Egocentric Action Recognition by Automatic Relation Modeling (TPAMI 2022 · similarity 0.89)
- Action Scene Graphs for Long-Form Understanding of Egocentric Videos (CVPR 2024 · similarity 0.89)
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.
