AI Paper Atlas

Research map / Detection and segmentation / Tracking

OVTrack: Open-Vocabulary Multiple Object Tracking

Siyuan Li, Tobias Fischer, Lei Ke, Henghui Ding, Martin Danelljan, Fisher Yu

CVPR 2023 · Detection and segmentation › Tracking · approach: Association cues for online MOT · 69 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 OVTrack instead of video-trained closed-set trackersTAO tracking · TAO validation, base and novel classesTETA: 35.5 vs 30.3; 27.8 vs 25.7Improved
Use OVTrack instead of video-trained closed-set trackersTAO tracking · TAO test, base and novel classesTETA: 32.6 vs 29.2; 24.1 vs 21.7Improved
Distill image embeddings in addition to text embeddings (ablation)OVTrack · TAO validationClsA: 15.6 -> 18.1Improved
Add image distillation to text distillation (ablation)OVTrack · TAO validationLocA: 50.5 -> 49.3Hurt
Add DDPM hallucination to standard augmentations (ablation)TETer-SwinT · TAO validationAssocA: 30.6 -> 33.0Improved
Add heavy augmentations to DDPM and standard augmentations (ablation)TETer-SwinT · TAO validationAssocA: 33.0 -> 35.5Improved
Add DDPM hallucination to standard augmentations (ablation)OVTrack · TAO validationAssocA: 31.1 -> 32.9Improved
Add heavy augmentations to DDPM and standard augmentations (ablation)OVTrack · TAO validationAssocA: 32.9 -> 35.7Improved
Use DDPM without language prompts or geometric transformation (ablation)OVTrack · TAO validationAssocA: 31.1 -> 30.6Hurt
Add geometric transformation before DDPM generation (ablation)OVTrack · TAO validationAssocA: 31.1 -> 32.4Improved
Add language prompting to DDPM with geometric transformation (ablation)OVTrack · TAO validationAssocA: 32.4 -> 32.9Improved
Transfer OVTrack trained on LVIS rather than target-dataset trainingBDD100K tracking · BDD100K validationTETA: 42.5 vs 36.1Improved

Later papers that compared with this method

11 later accepted papers report OVTrack as a baseline in their results tables; 9 of them report a better number on most of the results they share with it.

3 more not listed.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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