AI Paper Atlas

Research map / 3D vision / Face and shape

Scene Categorization From Contours: Medial Axis Based Salience Measures

Morteza Rezanejad, Gabriel Downs, John Wilder, Dirk B. Walther, Allan Jepson, Sven Dickinson, Kaleem Siddiqi

CVPR 2019 · 3D vision › Face and shape · approach: Primitive-based shape abstraction · 30 citations (OpenAlex)

Check 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 medial-axis salience-weighted contour channels to intact contoursPre-trained CNN features with linear SVM · Artist Scenestop-1 accuracy: Contours: 86.53; Contours, Ribbon, Separ.: 94.96Improved
Add ribbon-salience weighted contour channelPre-trained CNN features with linear SVM · MIT67top-1 accuracy: 26.36 -> 27.18 (VGG16); 23.88 -> 25.74 (Res50)Improved
Add taper-salience weighted contour channelPre-trained CNN features with linear SVM · Places365top-1 accuracy: 8.02 -> 11.73Improved
Add separation-salience weighted contour channelPre-trained CNN features with linear SVM · Places365top-1 accuracy: 8.02 -> 10.53Improved
Combine ribbon and separation salience channels with intact contoursPre-trained CNN features with linear SVM · Places365top-1 accuracy: 8.02 -> 14.23Improved
Replace the intact contour channel with all three salience-weighted channels (ablation)Pre-trained CNN features with linear SVM · MIT67top-1 accuracy: Contours 26.36; Ribbon, Taper, Separ. 25.84 (VGG16)Hurt
Use ribbon plus taper instead of ribbon plus separation salience (ablation)Contours plus salience channels, VGG16 · Artist Scenestop-1 accuracy: Contours, Ribbon, Taper 94.53; Contours, Ribbon, Separ. 94.96Hurt
Use ribbon plus taper instead of taper plus separation salience (ablation)Contours plus salience channels, ResNet50 · MIT67top-1 accuracy: Contours, Ribbon, Taper 25.44; Contours, Taper, Separ. 30.37Hurt
Use all three salience channels without intact contours (ablation)Pre-trained CNN features with linear SVM · Places365top-1 accuracy: Contours, Ribbon, Separ 14.23; Ribbon, Taper, Separ 12.64Hurt
Select the top 50% rather than bottom 50% of contour pixels by salience (ablation)CNN scene classification · Artist Scenes and MIT67top-1 accuracyImproved
Select the top 50% rather than bottom 50% of contour pixels by salience (ablation)Human scene categorization · Artist ScenesaccuracyImproved

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