AI Paper Atlas

Research map / Label-efficient and robust learning / Knowledge distillation

HEAD: HEtero-Assists Distillation for Heterogeneous Object Detectors

Luting Wang, Xiaojie Li, Yue Liao, Zeren Jiang, Jianlong Wu, Fei Wang, Chen Qian, Si Liu

ECCV 2022 · Label-efficient and robust learning › Knowledge distillation · approach: Knowledge distillation for detectors · 28 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 CKD loss to HEAD (AKD only) (ablation)HEAD, R18 RetinaNet, teacher R50 Faster R-CNN · COCO val, R18 RetinaNet student, teacher R50 Faster R-CNNmAP: 35.5 -> 36.2Improved
add CKD loss to HEAD (AKD only) (ablation)HEAD, MNV2 RetinaNet, teacher R50 Faster R-CNN · COCO val, MNV2 RetinaNet student, teacher R50 Faster R-CNNmAP: 32.5 -> 32.8Improved
add CKD loss to HEAD (AKD only) (ablation)HEAD, R18 RetinaNet, teacher R50 RepPoints · COCO val, R18 RetinaNet student, teacher R50 RepPointsmAP: 34.2 -> 34.3No clear effect
add assistant-based KD from R50 Faster R-CNNR18 RetinaNet · COCO val, R18 RetinaNet student, teacher R50 Faster R-CNNmAP: 31.7 -> 35.5Improved
use HEAD instead of homo-KD FitNet with RepPoints teacherR18 RetinaNet · COCO val, R18 RetinaNet student, teacher R50 RepPointsmAP: 31.5 -> 34.2Improved
apply homo-KD DeFeat with RepPoints teacherR18 RetinaNet · COCO val, R18 RetinaNet student, teacher R50 RepPointsmAP: 31.7 -> 30.9Hurt
use HEAD instead of G-DetKDR18 RetinaNet, teacher R50 Faster R-CNN · COCO val, R18 RetinaNet student, teacher R50 Faster R-CNNmAP: 35.4 -> 36.2Improved
use HEAD instead of G-DetKDR18 FCOS, teacher R50 Faster R-CNN · COCO val, R18 FCOS student, teacher R50 Faster R-CNNmAP: 34.1 -> 36.0Improved
use HEAD instead of G-DetKDMNV2 RetinaNet, teacher R50 Faster R-CNN · COCO val, MNV2 RetinaNet student, teacher R50 Faster R-CNNmAP: 32.6 -> 32.8No clear effect
add FCOS assistant to RetinaNet (teacher-free) (ablation)TF-HEAD, R18 RetinaNet · COCO val, TF-HEAD ablation, R18 RetinaNetmAP: 31.7 -> 32.9Improved
add CKD loss to FCOS assistant (ablation)TF-HEAD, R18 RetinaNet · COCO val, TF-HEAD ablation, R18 RetinaNetmAP: 32.9 -> 33.0No clear effect
add R-CNN assistant to RetinaNet (teacher-free) (ablation)TF-HEAD, R18 RetinaNet · COCO val, TF-HEAD ablation, R18 RetinaNetmAP: 31.7 -> 33.4Improved

Later papers that compared with this method

1 later accepted paper reports HEAD as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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