AI Paper Atlas

Research map / Detection and segmentation / Tracking

AnimalTrack: A Benchmark for Multi-Animal Tracking in the Wild

Libo Zhang, Junyuan Gao, Zhen Xiao, Heng Fan

IJCV 2022 · Detection and segmentation › Tracking · approach: Unified transformer tracking · 66 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 appearance features (DeepSORT) instead of motion-only association (SORT) (ablation)SORT · AnimalTrack test (association with Faster R-CNN detections)HOTA: 42.8% -> 38.2%Hurt
use quasi-dense matching association (QDTrack) instead of SORT (ablation)Faster R-CNN detections · AnimalTrack test (association with Faster R-CNN detections)HOTA: 42.8% -> 47.0%Improved
associate all detected boxes (ByteTrack) instead of SORT (ablation)Faster R-CNN detections · AnimalTrack test (association with Faster R-CNN detections)HOTA: 42.8% -> 36.3%Hurt
simple motion trackers SORT and IOUTrack versus recent learned trackersJDE, FairMOT, CTracker · AnimalTrack testHOTA: 42.8% HOTA vs 26.8%, 30.6%, 13.8%Improved
Transformer query-key tracking (TransTrack) instead of TrackformerTransformer trackers · AnimalTrack testHOTA: 31.0% -> 45.4%Improved
two-stage Faster R-CNN detection in Tracktor++Tracktor++ · AnimalTrack testMOTA: 55.2% MOTAImproved
dense region sampling in QDTrackQDTrack · AnimalTrack testMOTA: 55.7% MOTA, 56.3% IDF1Improved

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