Research map / Video / Action recognition
First Person Action Recognition Using Deep Learned Descriptors
Suriya Singh, Chetan Arora, C. V. Jawahar
Check 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 egocentric hand, head-motion, and saliency features to spatial and temporal streams | Three-stream action recognition · GTEA | accuracy: 59.47 -> 68.50 | Improved |
| add egocentric hand, head-motion, and saliency features to spatial and temporal streams | Three-stream action recognition · Kitchen | accuracy: 58.92 -> 66.23 | Improved |
| add egocentric hand, head-motion, and saliency features to spatial and temporal streams | Three-stream action recognition · ADL | accuracy: 34.20 -> 37.58 | Improved |
| add egocentric hand, head-motion, and saliency features to spatial and temporal streams | Three-stream action recognition · UTE | accuracy: 55.67 -> 60.17 | Improved |
| add head motion to hand masks (ablation) | Ego ConvNet (2D) · GTEA | accuracy: 51.76 -> 54.13 | Improved |
| add saliency map to hand masks and head motion (ablation) | Ego ConvNet (2D) · GTEA | accuracy: 54.13 -> 57.61 | Improved |
| add head motion to hand masks (ablation) | Ego ConvNet (3D) · GTEA | accuracy: 50.82 -> 53.15 | Improved |
| add saliency map to hand masks and head motion (ablation) | Ego ConvNet (3D) · GTEA | accuracy: 53.15 -> 55.79 | Improved |
| combine 2D and 3D Ego ConvNets (ablation) | Ego ConvNet · GTEA | accuracy: 57.61 and 55.79 -> 58.94 | Improved |
| add spatial and temporal TDD features (ablation) | Ego ConvNet (2D) · GTEA | accuracy: 57.61 -> 65.29 | Improved |
| add spatial and temporal TDD features (ablation) | Ego ConvNet (3D) · GTEA | accuracy: 55.79 -> 66.96 | Improved |
Closest papers by meaning
- Going Deeper into First-Person Activity Recognition (CVPR 2016 · similarity 0.93)
- Egocentric Action Recognition by Automatic Relation Modeling (TPAMI 2022 · similarity 0.89)
- Ego-Only: Egocentric Action Detection without Exocentric Transferring (ICCV 2023 · similarity 0.88)
- Identifying First-Person Camera Wearers in Third-Person Videos (CVPR 2017 · similarity 0.88)
- Learning to Recognize Actions on Objects in Egocentric Video With Attention Dictionaries (TPAMI 2021 · similarity 0.88)
- SOS! Self-Supervised Learning over Sets of Handled Objects in Egocentric Action Recognition (ECCV 2022 · similarity 0.88)
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.
