AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

Exploiting Lightweight Hierarchical ViT and Dynamic Framework for Efficient Visual Tracking

Ben Kang, Xin Chen, Jie Wen Zhao, Chunjuan Bo, Dong Wang, Huchuan Lu

IJCV 2025 · Architectures and efficiency › Attention and transformers · approach: Efficient and robust vision transformers · 5 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
fuse Max, Mid and Min features with the Bridge Module instead of using Max only (ablation)HiT-Base · LaSOT (ablation, 500 epochs)AUC: 62.1 -> 63.7Improved
add the deep Min feature to Max+Mid fusion (ablation)HiT-Base · LaSOT (ablation, 500 epochs)AUC: 62.6 -> 63.7Improved
add the shallow Max feature to Mid+Min fusion (ablation)HiT-Base · LaSOT (ablation, 500 epochs)AUC: 60.3 -> 63.7Improved
add the Mid feature to Max+Min fusion (ablation)HiT-Base · LaSOT (ablation, 500 epochs)AUC: 58.8 -> 63.7Improved
use Shrink Attention for downsampling instead of a stride-2 convolution (ablation)HiT-Base · LaSOT / TrackingNet / GOT-10kAUC: decreases of 5.6%, 2.8%, and 6.2% on LaSOT, TrackingNet, and GOT-10kImproved
apply training-free DyTracker acceleration (DyHiT-Route1 plus router) to OSTrack-256OSTrack-256 · LaSOTspeed and AUC: 70 fps -> 110 fps, 1.57x; AUC 69.1 -> 69.5Improved
apply training-free DyTracker acceleration to SeqTrack-B256SeqTrack-B256 · LaSOTspeed and AUC: 2.68x speedup; AUC 69.9 -> 69.9Improved
apply training-free DyTracker acceleration to TransTTransT · TrackingNet / LaSOT / GOT-10kaveraged AUC change and speed: -0.50 average change, 1.86x speedMixed
apply training-free DyTracker acceleration to STARK-ST101STARK-ST101 · TrackingNet / LaSOT / GOT-10kaveraged AUC change and speed: +0.03 average change, 2.93x speedImproved
apply training-free DyTracker acceleration to OSTrack-256OSTrack-256 · UAV123AUC: 68.3 -> 70.8Improved
dynamic routing with router threshold 0.75 instead of full model only (threshold 1) (ablation)DyHiT · VOT2021 real-timeEAO: 0.252 -> 0.253No clear effect
use only Route1 (threshold 0) instead of full model (ablation)DyHiT · VOT2021 real-timeEAO: 0.252 -> 0.250Hurt

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