atlas

Research map / Detection and segmentation / CNN features

Real-Time Rotation-Invariant Face Detection With Progressive Calibration Networks

CVPR 2018 · Detection and segmentation › CNN features · approach: Box regression and loss design · 110 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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
Use progressive coarse-to-fine orientation calibration with staged face classificationPCN · Multi-Oriented FDDB (Ave)recall rate at 100 FP (%): 84.9 -> 87.4Helped
Use progressive coarse-to-fine orientation calibration instead of a rotation routerPCN · Multi-Oriented FDDB (Ave)recall rate at 100 FP (%): 84.8 -> 87.4Helped
Use progressive coarse-to-fine orientation calibration instead of four rotated detector passesPCN · Multi-Oriented FDDB (Ave)recall rate at 100 FP (%): 85.5 -> 87.4Helped
Use progressive coarse-to-fine orientation calibration instead of Cascade CNN + STNPCN · Multi-Oriented FDDB (Ave)recall rate at 100 FP (%): 85.5 -> 87.4Helped
Use progressive calibration rather than direct continuous angle regression in the routerRotation Router · Rotation calibrationmean angular errorHelped
Add a coarse binary orientation-classification stagePCN calibration · PCN-1 orientation classificationaccuracyHelped
Add a coarse ternary orientation-classification stagePCN calibration · PCN-2 orientation classificationaccuracyHelped

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