Research map / Video / Video segmentation and tracking
Joint Person Segmentation and Identification in Synchronized First- and Third-person Videos
Mingze Xu, Chenyou Fan, Yuchen Wang, Michael S. Ryoo, David J. Crandall
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 optical flow stream to image stream (ablation) | two-stream joint FCN · IU ShareView third-third | IoU: 49.3 -> 62.7 | Improved |
| add optical flow stream to image stream (ablation) | two-stream joint FCN · IU ShareView third-first | ACC: 52.7 -> 73.1 | Improved |
| add image stream to optical flow stream (ablation) | two-stream joint FCN · IU ShareView third-first | mAP: 55.1 -> 65.2 | Improved |
| add soft attention re-weighting of features (ablation) | third-third joint model · IU ShareView third-third | ACC: 48.9 -> 55.5 | Improved |
| add soft attention re-weighting of features (ablation) | third-first joint model · IU ShareView third-first | ACC: 61.7 -> 73.1 | Improved |
| add soft attention re-weighting of features (ablation) | third-first joint model · UTokyo Ego-Surf | ACC: 44.0 -> 55.2 | Improved |
| add soft attention re-weighting of features (ablation) | third-third joint model · UTokyo Ego-Surf | ACC: 36.7 -> 42.0 | Improved |
| add soft attention re-weighting of features (ablation) | third-third joint model · IU ShareView third-third | IoU: 60.6 -> 62.7 | Improved |
| replace ground-truth bounding-box cropping with soft attention from predicted segmentation | joint model · IU ShareView third-third | mAP: 44.2 -> 49.0 | Improved |
| replace ground-truth bounding-box cropping with soft attention from predicted segmentation | joint model · IU ShareView third-first | mAP: 64.1 -> 65.2 | Improved |
| jointly learn matching with segmentation instead of segmentation only | two-stream FCN · IU ShareView third-third | IoU: 57.3 -> 62.7 | Improved |
| jointly learn matching with segmentation instead of segmentation only | two-stream FCN · IU ShareView third-first | IoU: 57.3 -> 61.9 | Improved |
Closest papers by meaning
- Identifying First-Person Camera Wearers in Third-Person Videos (CVPR 2017 · similarity 0.89)
- Fusing Personal and Environmental Cues for Identification and Segmentation of First-Person Camera Wearers in Third-Person Views (CVPR 2024 · similarity 0.89)
- Learning from Synchronization: Self-Supervised Uncalibrated Multi-View Person Association in Challenging Scenes (CVPR 2025 · similarity 0.87)
- Integrating Egocentric Videos in Top-view Surveillance Videos: Joint Identification and Temporal Alignment (ECCV 2018 · similarity 0.87)
- Human Detection and Segmentation via Multi-View Consensus (ICCV 2021 · similarity 0.86)
- MultiPly: Reconstruction of Multiple People from Monocular Video in the Wild (CVPR 2024 · 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.
