Research map / Detection and segmentation / Open-set and few-shot detection
EgoObjects: A Large-Scale Egocentric Dataset for Fine-Grained Object Understanding
Chenchen Zhu, Fanyi Xiao, Andres Alvarado, Yasmine Babaei, Jiabo Hu, Hichem El-Mohri, Sean Culatana, Roshan Sumbaly, Zhicheng Yan
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 |
|---|---|---|---|
| Use target-aware detector TA-IDet instead of target-agnostic RPN+SFNet | instance detection · EgoObjects test split (R50) | AP50: 25.4 -> 35.2 | Improved |
| Add index loss (ablation) | TA-IDet · EgoObjects val | AP: 13.6 -> 18.7 | Improved |
| Add negative loss (ablation) | TA-IDet · EgoObjects val | AP: 17.4 -> 18.7 | Improved |
| Use 5x5 instead of 1x1 T_cls feature resolution (ablation) | TA-IDet · EgoObjects val | AP: 11.3 -> 18.7 | Improved |
| Increase T_cls resolution from 5 to 7 (ablation) | TA-IDet · EgoObjects val | AP: 18.7 -> 18.3 | No clear effect |
| Use single-stage FCOS instead of two-stage FasterRCNN | category detection · EgoObjects val | AP75: 23.6 -> 26.4 (R101) | Improved |
| Use DeformDETR-Swin backbone instead of CNN detectors | category detection · EgoObjects val | AP: 24.2 -> 33.5 | Improved |
| Evaluate on unseen-category instances instead of seen ones | TA-IDet · EgoObjects instance detection | AP50: 35.0 -> 21.7 (val R50) | Hurt |
Closest papers by meaning
- Self-Supervised Object Detection from Egocentric Videos (ICCV 2023 · similarity 0.94)
- EgoTracks: A Long-term Egocentric Visual Object Tracking Dataset (NeurIPS 2023 · similarity 0.91)
- Towards Stable Self-Supervised Object Representations in Unconstrained Egocentric Video (CVPR 2026 · similarity 0.90)
- ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models (NeurIPS 2019 · similarity 0.89)
- Robust Egocentric Referring Video Object Segmentation via Dual-Modal Causal Intervention (NeurIPS 2025 · similarity 0.88)
- Learning to Segment Referred Objects from Narrated Egocentric Videos (CVPR 2024 · similarity 0.87)
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.
