AI Paper Atlas

Research map / Detection and segmentation / CNN features

Higher Order Conditional Random Fields in Deep Neural Networks

Anurag Arnab, Sadeep Jayasumana, Shuai Zheng, Philip H. S. Torr

ECCV 2016 · Detection and segmentation › CNN features · approach: Context modeling for semantic segmentation · 212 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
add detection higher order potentials (ablation)Pairwise CRF-RNN · VOC 2012 reduced valmean IoU (%): 72.9 -> 74.9Improved
add superpixel higher order potentials (ablation)Pairwise CRF-RNN · VOC 2012 reduced valmean IoU (%): 72.9 -> 74.0Improved
add both detection and superpixel potentials (ablation)Pairwise CRF-RNN · VOC 2012 reduced valmean IoU (%): 72.9 -> 75.8Improved
add superpixel potentials on top of detection potentials (ablation)Detections-only CRF · VOC 2012 reduced valmean IoU (%): 74.9 -> 75.8Improved
train CRF end-to-end instead of piecewise (ablation)Higher Order CRF (FCN-8s) · VOC 2012 reduced valmean IoU (%): 73.6 -> 75.8Improved
train CRF end-to-end instead of piecewise (ablation)Higher Order CRF (DCN) · VOC 2012 reduced valmean IoU (%): 73.5 -> 75.0Improved
train pairwise CRF end-to-end instead of piecewise (ablation)Pairwise CRF (FCN-8s) · VOC 2012 reduced valmean IoU (%): 69.5 -> 72.9Improved
add higher order potentials to pairwise CRF with DCN unariesPairwise CRF (DCN) · VOC 2012 reduced valmean IoU (%): 72.5 -> 75.0Improved
use latent-variable detection potentials instead of adding detection mask to unaries (ablation)Detection unary baseline · VOC 2012 reduced valmean IoU (%): 73.6 -> 74.9Improved
add higher order potentials to CRF-RNNCRF-RNN · VOC 2012 testmean IoU (%): 74.7 -> 77.9Improved
add higher order potentials to CRF-RNNCRF-RNN · PASCAL Context val (59 classes)mean IoU (%): 39.3 -> 41.3Improved
add superpixel potentials (ablation)Pairwise CRF-RNN · VOC 2012 reduced val (boundary IoU)boundary IoUHurt

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