AI Paper Atlas

Research map / Detection and segmentation / CNN features

SelfMOTR: Revisiting MOTR with Self-Generating Detection Priors

Fabian Gülhan, Emil Mededovic, Yuli Wu, Johannes Stegmaier

ECCV 2026 · Detection and segmentation › CNN features · approach: Efficient DETR query design · 0 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 self-generated proposals (decoupled proposal pass) (ablation)MOTR · DanceTrack valHOTA: 51.2 -> 58.2; AssA 38.4 -> 47.4Improved
add anchor proposals from detection-pretrained MOTR (ablation)MOTR · DanceTrack valHOTA: 51.2 -> 58.0Improved
add detection pretraining (ablation)MOTR · DanceTrack valHOTA: 51.2 -> 51.5; MOTA 74.4 -> 73.8No clear effect
add query pretraining (frozen detect queries) (ablation)MOTR · DanceTrack valAssA: 38.4 -> 41.1; IDF1 49.1 -> 51.9Improved
add detection distillation from frozen teacher (ablation)MOTR · DanceTrack valDetA: 68.8 -> 72.0; MOTA 74.4 -> 80.1; AssA 38.4 -> 37.9Mixed
use self proposal instead of anchor proposals from pretrained MOTR (ablation)SelfMOTR · DanceTrack valHOTA: 58.0 -> 58.2No clear effect
use 4D anchor proposals instead of learnable 4D anchors (ablation)MOTR · DanceTrack valHOTA: 55.8 -> 58.0Improved
add HSV augmentation and CrowdHuman joint training (ablation)SelfMOTR · DanceTrack valHOTA: 58.2 -> 64.3Improved
add denoising stage (ablation)SelfMOTR · DanceTrack valHOTA: 64.3 -> 64.4No clear effect
share decoder weights between proposal and tracking passes (ablation)SelfMOTR · DanceTrack testHOTA: 66.2 -> 69.2; AssA 54.5 -> 59.3Improved
lower proposal confidence threshold from 0.50 to 0.05 (ablation)SelfMOTR · DanceTrack valHOTA: 57.7 -> 58.2Improved
add track queries to MOTR detection (ablation)MOTR · DanceTrack valAP: 66.3 -> 60.0Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 5 on most shared results.

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