AI Paper Atlas

Research map / Detection and segmentation / Tracking

Cannot See the Forest for the Trees: Aggregating Multiple Viewpoints To Better Classify Objects in Videos

Sukjun Hwang, Miran Heo, Seoung Wug Oh, Seon Joo Kim

CVPR 2022 · Detection and segmentation › Tracking · approach: Unified transformer tracking · 4 citations (OpenAlex)

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
Add tracklet-level set classification to QDTrackQDTrack · TAO validationTrackAP50: 15.8 -> 19.9Improved
Add tracklet-level set classification to QDTrackQDTrack · TAO testTrackAP50: 12.4 -> 15.7Improved
Add tracklet-level set classification to QDTrackQDTrack · YouTube-VIS 2019AP: 34.4 -> 37.7Improved
Allow tracklets to mix multiple identities while still restricting them to one class (ablation)Set classifier · TAO validationTrackAP50: 17.7 -> 17.2Hurt
Allow multiple classes in multi-identity tracklets (ablation)Set classifier · TAO validationTrackAP50: 17.2 -> 19.9Improved
Increase generated tracklets per batch from 32 to 256 (ablation)Set classifier · TAO validationTrackAP50: 17.7 -> 19.9Improved
Add video-derived tracklet training to image-derived tracklet training (ablation)Set classifier · TAO validationTrackAP50: 17.1 -> 19.9Improved
Train on videos only instead of images and videos (ablation)Set classifier · TAO validationTrackAP50: 19.9 -> 6.5Hurt
Use tracklets of length [16, 32) rather than [8, 16) or [32, 64) (ablation)Set classifier · TAO validationTrackAP50: 19.0, 18.3 -> 19.9Improved
Use tail-favoring RoI sampling with p=0.5 rather than uniform sampling (ablation)Set classifier · TAO validationTrackAP50: 15.6 -> 19.9Improved
Add instance-wise loss (ablation)Set classifier · TAO validationTrackAP50: 18.7 -> 18.9Improved
Add clustering loss alongside instance-wise loss (ablation)Set classifier · TAO validationTrackAP50: 18.9 -> 19.9Improved

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