Research map / 3D vision / Human pose estimation
Recognizing Activities of Daily Living With a Wrist-Mounted Camera
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 spatial weighting to LCD descriptors (ablation) | LCD on MILADL WCD · MILADL WCD | mean classification accuracy: 78.6 -> 82.0 | Helped |
| Add temporal weighting to spatially weighted descriptors (ablation) | DSAR on MILADL WCD · MILADL WCD | mean classification accuracy: 82.0 -> 83.7 | Helped |
| Apply spatial and temporal weighting (ablation) | LCD on MILADL HCD · MILADL HCD | mean classification accuracy: 62.4 -> 62.0 | Hurt |
| Add spatial pyramid pooling to LCD (ablation) | LCD on MILADL WCD · MILADL WCD | mean classification accuracy: 78.6 -> 73.4 | Hurt |
| Use optimized spatial and temporal weights instead of an unweighted spatiotemporal pyramid (ablation) | DSTAR on MILADL WCD · MILADL WCD | mean classification accuracy: 77.0 -> 83.7 | Helped |
| Add spatial and temporal aggregation to LCD (ablation) | LCD on UCIADL · UCIADL | mean classification accuracy: 73.7 -> 72.6 | Hurt |
| Add spatial and temporal aggregation to LCD (ablation) | LCD on UCF101 · UCF101 | mean classification accuracy: 76.8 -> 79.3 | Helped |
| Fuse iDT motion features with DSTAR (ablation) | DSTAR on MILADL WCD · MILADL WCD | mean classification accuracy: 83.7 -> 85.5 | Helped |
| Use DSTAR from WCD with iDT from HCD instead of a single-camera feature configuration (ablation) | MILADL camera fusion · MILADL | mean classification accuracy: 89.7 | Helped |
| Use a wrist-mounted rather than head-mounted camera with LCD | LCD on MILADL · MILADL WCD versus HCD | mean classification accuracy: 62.4 -> 78.6 | Helped |
Closest papers by meaning
- Seeing Invisible Poses: Estimating 3D Body Pose From Egocentric Video (CVPR 2017 · similarity 0.83)
- HARDVS: Revisiting Human Activity Recognition with Dynamic Vision Sensors (AAAI 2024 · similarity 0.83)
- EgoSim: An Egocentric Multi-view Simulator and Real Dataset for Body-worn Cameras during Motion and Activity (NeurIPS 2024 · similarity 0.82)
- Activity-Biometrics: Person Identification from Daily Activities (CVPR 2024 · similarity 0.82)
- ZeroHAR: Sensor Context Augments Zero-Shot Wearable Action Recognition (AAAI 2025 · similarity 0.81)
- EgoLife: Towards Egocentric Life Assistant (CVPR 2025 · similarity 0.81)
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.