AI Paper Atlas

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

Distilling Object Detectors with Feature Richness

Zhixing Du, Rui Zhang, Ming Chang, Xishan Zhang, Shaoli Liu, Tianshi Chen, Yunji Chen

NeurIPS 2021 · Label-efficient and robust learning › Knowledge distillation · approach: Knowledge distillation for detectors · 36 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
Apply FRS distillation to RetinaNet ResNet-50RetinaNet ResNet-50 with ResNet-101 teacher · COCOmAP: 37.4 -> 39.7Improved
Apply FRS distillation to GFL ResNet-50 with the 1x scheduleGFL ResNet-50 with ResNet-101 teacher · COCOmAP: 40.2 -> 43.6Improved
Apply FRS distillation to GFL ResNet-50 with the 2x scheduleGFL ResNet-50 with ResNet-101 teacher · COCOmAP: 42.9 -> 44.7Improved
Apply FRS distillation to Faster R-CNN ResNet-50Faster R-CNN ResNet-50 with ResNet-101 teacher · COCOmAP: 37.4 -> 39.5Improved
Apply FRS distillation to FCOS ResNet-50FCOS ResNet-50 with ResNet-101 teacher · COCOmAP: 38.5 -> 40.9Improved
Add FPN feature distillation (ablation)RetinaNet ResNet-50 FRS · COCOmAP: 37.4 -> 39.4Improved
Add classification-head distillation (ablation)RetinaNet ResNet-50 FRS · COCOmAP: 37.4 -> 38.4Improved
Add both FPN and classification-head distillation (ablation)RetinaNet ResNet-50 FRS · COCOmAP: 37.4 -> 39.7Improved
Add high-richness false-positive background regions to TP distillation (ablation)RetinaNet ResNet-50 FPN distillation · COCOmAP: 39.1 -> 39.4Improved
Add false-negative foreground regions to TP distillation (ablation)RetinaNet ResNet-50 FPN distillation · COCOmAP: 39.1 -> 39.0Hurt
Add true-negative background regions to TP distillation (ablation)RetinaNet ResNet-50 FPN distillation · COCOmAP: 39.1 -> 38.7Hurt

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