AI Paper Atlas

Research map / Detection and segmentation / CNN features

Training Deep Networks to Be Spatially Sensitive

Nicholas Kolkin, Eli Shechtman, Gregory Shakhnarovich

ICCV 2017 · Detection and segmentation › CNN features · approach: Local features and context in CNNs · 5 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
replace cross-entropy with approximate weighted F-measure loss (ablation)SZN · MSRA-BF^w_1: 0.829 -> 0.856Improved
replace cross-entropy with approximate weighted F-measure loss (ablation)SZN · HKU-ISF^w_1: 0.788 -> 0.826Improved
replace cross-entropy with approximate weighted F-measure loss (ablation)SZN · ECSSDF^w_1: 0.796 -> 0.827Improved
replace cross-entropy with approximate weighted F-measure loss (ablation)SZN · PASCAL-SF^w_1: 0.657 -> 0.680Improved
replace cross-entropy with approximate weighted F-measure loss (ablation)SZN · ECSSDMAE: 0.069 -> 0.073No clear effect
replace cross-entropy with approximate weighted F-measure loss (ablation)SZN · Portrait segmentationMIoU: 96.53 -> 97.13Improved
add the foreground reweighting matrix A (ablation)Proposed loss · MSRA-B objective comparisonF^w_1: 0.836 -> 0.856Improved
add the false-positive boundary weighting B (ablation)Proposed loss · MSRA-B objective comparisonF^w_1: 0.835 -> 0.856Improved
use proposed loss instead of CE with 2x foreground weight (ablation)SZN · MSRA-B objective comparisonAUROC: 0.976 -> 0.988Improved
use proposed loss instead of CE ignoring 3-pixel border band (ablation)SZN · MSRA-B objective comparisonF^w_1: 0.807 -> 0.856Improved
use proposed loss instead of CE with 2x background weight (ablation)SZN · MSRA-B objective comparisonF^w_1: 0.797 -> 0.856Improved
approximate F^w_beta with convolutionsF^w_beta metric · 15 ECSSD imagescomputation time: 37 minutes -> 0.33 secondsImproved

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