AI Paper Atlas

Research map / Detection and segmentation / Tracking

Rethinking Temporal Modeling in Visual Object Tracking via Decoupled Auxiliary Supervision

Dailing Zhang, Shiyu Hu, Honghao Fu, Xiaokun Feng, Yipei Wang, Kang Hao Cheong, Kaiqi Huang

ECCV 2026 · Detection and segmentation › Tracking · approach: Transformer-based video object tracking · 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 decoupled auxiliary supervision (template-masked auxiliary pass) in training (ablation)HiViT tracker · LaSOTAUC: 73.0 -> 73.4Improved
Add decoupled auxiliary supervision (ablation)ViT tracker · LaSOTAUC: 71.0 -> 71.4Improved
Add temporal tokens with standard joint training (ablation)HiViT tracker · LaSOTAUC: 72.8 -> 73.0No clear effect
Perturb temporal tokens at inferenceStandard temporal trackers (ARTrackV2, EVPTrack, AQATrack, ODTrack) · LaSOTAUC: 71.4 -> 71.2 (ARTrackV2)No clear effect
Provide ground-truth cropped template or GT-centered search at inferenceSUTrack / OSTrack · LaSOTAUC: 73.1 -> 82.1; 69.9 -> 77.9Improved
Give auxiliary branch template access (ablation)DAS · LaSOTAUC: 73.4 -> 72.6Hurt
Use adaptive (learnable-uncertainty) loss weight instead of fixed (ablation)DAS · LaSOTAUC: 73.2 -> 73.4Improved
Use LoRA rank 64 vs no LoRA (ablation)DAS · LaSOTAUC: 73.2 -> 73.4Improved
Increase LoRA rank to 128 (ablation)DAS · LaSOTAUC: 73.4 -> 72.7Hurt
Use joint instead of sequential optimization of auxiliary task (ablation)DAS · LaSOTAUC: 72.7 -> 73.4Improved
Add DAS to training (ablation)Temporal-token gradient ratio · LaSOTR_H: 0.165 -> 0.624Improved
Use DAS (training-only) at inferenceDASTrack vs SUTrack · Speed/paramsfps: 34.0 -> 34.0No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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