AI Paper Atlas

Research map / Graph learning / Knowledge and temporal graphs

Layout-Graph Reasoning for Fashion Landmark Detection

Weijiang Yu, Xiaodan Liang, Ke Gong, Chenhan Jiang, Nong Xiao, Liang Lin

CVPR 2019 · Graph learning › Knowledge and temporal graphs · approach: Graph reasoning for visual recognition · 41 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 hierarchical layout-graph reasoning with four stacked LGR layersVGG16 · FLDAvg. NE: .0871 -> .0419Improved
add hierarchical layout-graph reasoning with four stacked LGR layersVGG16 · DeepFashionAvg. NE: .0484 -> .0336Improved
use five instead of four stacked LGR layers (ablation)LGR · FLDAvg. NE: .0419 -> .0405Improved
replace LGR clustering/deconvolution with one normal graph layer (ablation)VGG16 on FLD · FLDAvg. NE: .0419 -> .0531Hurt
replace LGR clustering/deconvolution with normal graph layers (ablation)VGG16 on DeepFashion · DeepFashionAvg. NE: .0336 -> .0437Hurt
increase normal graph depth from two to eight layers (ablation)normal graph model on FLD · FLDAvg. NE: .0471 -> .0954Hurt
increase normal graph depth from two to eight layers (ablation)normal graph model on DeepFashion · DeepFashionAvg. NE: .0437 -> .0779Hurt
use three instead of one graph clustering/deconvolution operation (ablation)LGR · FLDAvg. NE: .0488 -> .0419Improved
use three instead of one graph clustering/deconvolution operation (ablation)LGR · DeepFashionAvg. NE: .0403 -> .0336Improved
inject four stacked LGR layers before ConvBlock5 rather than ConvBlock3 (ablation)VGG16 on FLD · FLDAvg. NE: .0574 -> .0419Improved
inject four stacked LGR layers before ConvBlock5 rather than ConvBlock1 (ablation)VGG16 on FLD · FLDAvg. NE: .0811 -> .0419Improved
add hierarchical layout-graph reasoningVGG16 · FFLD testAvg. NE: .1272 -> .1180Improved

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