AI Paper Atlas

Research map / Detection and segmentation / CNN features

Relation Distillation Networks for Video Object Detection

Jiajun Deng, Yingwei Pan, Ting Yao, Wengang Zhou, Houqiang Li, Tao Mei

ICCV 2019 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 249 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 within-frame object relation module (ablation)Faster R-CNN · ImageNet VID valmAP (%): 75.4 -> 78.5Improved
add cross-frame relation in basic stage (ablation)Faster R-CNN · ImageNet VID valmAP (%): 75.4 -> 80.9Improved
add advanced stage relation distillation (ablation)RDN basic stage · ImageNet VID valmAP (%): 80.9 -> 81.8Improved
replace within-frame relation with cross-frame (spatio-temporal) relation (ablation)Faster R-CNN + relation · ImageNet VID valmAP (%): 78.5 -> 80.9Improved
add Box Linking with Relations post-processingRDN (ResNet-101) · ImageNet VID valmAP (%): 81.8 -> 83.8Improved
add Box Linking with Relations post-processingRDN (ResNeXt-101-64x4d) · ImageNet VID valmAP (%): 83.2 -> 84.7Improved
use BLR instead of Tube Linking (ablation)RDN (ResNet-101) · ImageNet VID valmAP (%): 83.4 -> 83.8Improved
use stronger backbone ResNeXt-101-64x4d instead of ResNet-101RDN · ImageNet VID valmAP (%): 81.8 -> 83.2Improved
increase temporal spanning range T from 3 to 18 (ablation)RDN · ImageNet VID valmAP (%): 80.3 -> 81.8Improved
increase T beyond 18 to 24 (ablation)RDN · ImageNet VID valmAP (%): 81.8 -> 81.7No clear effect
increase relation modules in basic stage from 0 to 2 (ablation)RDN basic stage · ImageNet VID valmAP (%): 75.4 -> 80.9Improved
stack more than two relation modules (N_b=4) (ablation)RDN basic stage · ImageNet VID valmAP (%): 80.9 -> 80.4Hurt

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