AI Paper Atlas

Research map / Graph learning / GNN architectures

Efficient Graph Cut Optimization for Full CRFs with Quantized Edges

Olga Veksler

TPAMI 2019 · Graph learning › GNN architectures · approach: Theory of graph convolutional networks · 18 citations (OpenAlex)

Read the paper

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
use expansion graph cut instead of mean fieldQuantized-edge Full CRF · Multilabel Full CRF energyrelative energy increaseImproved
use graph cut instead of mean field at large smoothnessBinary Full CRF · lambda=0.1 vs 2relative energy increase: 5.3% -> 40% (mean field)Improved
add Full CRF optimization to unary termsCNN unary · PASCAL VOC 2012 testIOU: 67.143 -> 67.7484Improved
add Full CRF optimization on top of superpixel labeling (ablation)superpixel labeling · PASCAL VOC 2012 testIOU: 65.8899 -> 67.7484Improved
use our optimization instead of mean fieldQuantized-edge Full CRF · PASCAL VOC 2012 testIOU: 67.3 -> 67.7484Improved
use our method instead of mean fieldQuantized-edge Full CRF · Binary Full CRF running timerunning time (s): 0.012 -> 0.31Hurt
use our method instead of exact graph cutBinary Full CRF · 70x70 imagesrunning time (s): 7.1 -> 0.31Improved
use our method instead of superpixel ICM at large lambdaMultilabel Full CRF · larger lambdarelative energy increaseImproved

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