AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

DARTH: Holistic Test-time Adaptation for Multiple Object Tracking

Mattia Segu, Bernt Schiele, Fisher Yu

ICCV 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Domain-adaptive object detection · 17 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
add DARTH test-time adaptation to QDTrack under sim-to-real shiftQDTrack · SHIFT -> BDD100K (average)MOTA: -66.4 -> 8.3Improved
apply patch contrastive learning alone without detection consistency (ablation)DARTH · SHIFT -> BDD100K (average)DetA: 12.0 -> 9.4Hurt
add detection consistency alone (ablation)DARTH · SHIFT -> BDD100K (average)MOTA: -66.4 -> -37.6Improved
add EMA teacher updates to detection consistency (ablation)DARTH · SHIFT -> BDD100K (average)MOTA: -37.6 -> 6.1Improved
add patch contrastive learning to EMA + DC (ablation)DARTH · SHIFT -> BDD100K (average)MOTA: 6.1 -> 8.3Improved
generate student view with photometric augmentation from teacher view (ablation)DARTH · SHIFT -> BDD100K (average)MOTA: -5.8 -> 8.3Improved
add photometric augmentation to the teacher view (ablation)DARTH · SHIFT -> BDD100K (average)MOTA: -5.8 -> -6.8Hurt
use Tent (entropy minimization on BN) for adaptationQDTrack · SHIFT -> BDD100K (average)MOTA: -66.4 -> 0.0Hurt
use SFOD detector-only self-training for adaptationQDTrack · MOT17 (+CH) -> BDD100KMOTA: 28.3 -> 17.0Hurt
add DARTH to MOT17 (+CH) source model on BDD100KQDTrack · MOT17 (+CH) -> BDD100KMOTA: 28.3 -> 23.4Hurt
pretrain detector on CrowdHuman before training QDTrack on MOT17 (ablation)QDTrack source model · MOT17 -> DanceTrack/BDD100KDetA: up to +9.4 DetA and +6.8 AssAImproved
DARTH versus SFOD on BDD100K -> MOT17 detectionDARTH · BDD100K -> MOT17DetA: 29.6 (SFOD) vs 29.4Mixed

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