AI Paper Atlas

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

ICCV 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Open-set and unknown object detection · 29 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use target-aware detector TA-IDet instead of target-agnostic RPN+SFNetinstance detection · EgoObjects test split (R50)AP50: 25.4 -> 35.2Improved
Add index loss (ablation)TA-IDet · EgoObjects valAP: 13.6 -> 18.7Improved
Add negative loss (ablation)TA-IDet · EgoObjects valAP: 17.4 -> 18.7Improved
Use 5x5 instead of 1x1 T_cls feature resolution (ablation)TA-IDet · EgoObjects valAP: 11.3 -> 18.7Improved
Increase T_cls resolution from 5 to 7 (ablation)TA-IDet · EgoObjects valAP: 18.7 -> 18.3No clear effect
Use single-stage FCOS instead of two-stage FasterRCNNcategory detection · EgoObjects valAP75: 23.6 -> 26.4 (R101)Improved
Use DeformDETR-Swin backbone instead of CNN detectorscategory detection · EgoObjects valAP: 24.2 -> 33.5Improved
Evaluate on unseen-category instances instead of seen onesTA-IDet · EgoObjects instance detectionAP50: 35.0 -> 21.7 (val R50)Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy