AI Paper Atlas

Research map / Architectures and efficiency / Attention and transformers

HiFT: Hierarchical Feature Transformer for Aerial Tracking

Ziang Cao, Changhong Fu, Junjie Ye, Bowen Li, Yiming Li

ICCV 2021 · Architectures and efficiency › Attention and transformers · approach: Efficient image transformers · 355 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 original transformer with object queries (ablation)Baseline · UAV20Lprecision: 0.611 -> 0.597Hurt
add original transformer with object queries (ablation)Baseline · UAV20Lsuccess: 0.463 -> 0.446Hurt
replace object queries with feature map in transformer (FT) (ablation)Baseline · UAV20Lprecision: 0.611 -> 0.675Improved
replace object queries with feature map in transformer (FT) (ablation)Baseline · UAV20Lsuccess: 0.463 -> 0.496Improved
add modulation layer to feature transformer (FT to HFT) (ablation)Baseline+FT · UAV20Lprecision: 0.675 -> 0.763Improved
add modulation layer to feature transformer (FT to HFT) (ablation)Baseline+FT · UAV20Lsuccess: 0.496 -> 0.566Improved
omit positional encoding on M5 (vs direct PE) (ablation)HiFT · UAV20Lprecision: 0.689 -> 0.763Improved
omit positional encoding on M5 (vs direct PE) (ablation)HiFT · UAV20Lsuccess: 0.523 -> 0.566Improved
use circular classification label instead of rectangle label (ablation)HiFT · UAV20Lprecision: 0.629 -> 0.763Improved
use circular classification label instead of rectangle label (ablation)HiFT · UAV20Lsuccess: 0.486 -> 0.566Improved
HiFT (AlexNet) vs SiamRPN++ (ResNet-50) deeper backboneHiFT · Four aerial benchmarks averageavg precision: 0.774 -> 0.783Improved
HiFT vs SiamRPN++ on UAV20L precisionSiamRPN++ · UAV20Lprecision: 0.696 -> 0.763Improved

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