AI Paper Atlas

Research map / Detection and segmentation / CNN features

Recurrent Scale Approximation for Object Detection in CNN

Yu Liu, Hongyang Li, Junjie Yan, Fangyin Wei, Xiaogang Wang, Xiaoou Tang

ICCV 2017 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 98 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 RSA unit to LRN for approximating smaller-scale feature maps (ablation)LRN · AFW/FDDB/MALF averagetest runtime (ms per image): 28.78 -> 14.50Improved
use LRN+RSA instead of single-anchor RPN baselineRPN_s · FDDBrecall at 1% fpi (%): 90.61% -> 91.92%Improved
use LRN+RSA instead of single-anchor RPN baselineRPN_s · MALFrecall at 1% fpi (%): 88.81% -> 90.09%Improved
use LRN+RSA instead of single-anchor RPN baselineRPN_s · AFWAP (%): 99.90% -> 99.96%Improved
use single-anchor multi-shot design instead of multi-anchor RPNRPN_m · FDDBrecall at 1% fpi (%): 86.89% -> 90.61%Improved
use LRN+RSA instead of multi-anchor RPNRPN_m · MALFrecall at 1% fpi (%): 84.65% -> 90.09%Improved
plug RSA at deeper layer (final feature, after res3c) instead of res2b (ablation)RSA unit · validation seterror rate: 3.44% -> almost 100%Hurt
roll out RSA five times instead of once (ablation)RSA unit at res2b · validation seterror rate: 3.44% -> 5.9%Hurt
supervise landmark features with regression predictions instead of ground-truth targets (ablation)LRN · face detectionperformanceImproved
use RSA+RPN instead of original 18-anchor RPNoriginal RPN · ILSVRC DET val2 (100 proposals)recall (%): 88.7 -> 89.1Improved
use RSA+RPN instead of single-scale RPNsingle-scale RPN · ILSVRC DET val2speed (ms per image): 249 ms -> 124 msImproved
use RSA+RPN instead of original RPNoriginal RPN · ILSVRC DET val2speed (ms per image): 158 ms -> 124 msImproved

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