Revisiting Deep Structured Models for Pixel-Level Labeling with Gradient-Based Inference

DOI10.1137/18M1167267zbMath1448.68442OpenAlexW2899901372WikidataQ125260311 ScholiaQ125260311MaRDI QIDQ5230425

Måns Larsson, Anurag Arnab, Shuai Zheng, Philip H. S. Torr, Fredrik Kahl

Publication date: 22 August 2019

Published in: SIAM Journal on Imaging Sciences (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1137/18m1167267

zbMATH Keywords

deep learningconditional random fieldssemantic segmentationdeep structured models

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Computing methodologies for image processing (68U10) Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Machine vision and scene understanding (68T45)

Uses Software

• DeepLab
• GrabCut
• Caffe
• MS-COCO
• PASCAL VOC
• RefineNet
• Faster R-CNN

Cites Work

• Pseudo-Boolean optimization
• Filter-based mean-field inference for random fields with higher-order terms and product label-spaces
• Mirror descent and nonlinear projected subgradient methods for convex optimization.
• Efficient Relaxations for Dense CRFs with Sparse Higher-Order Potentials
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: Revisiting Deep Structured Models for Pixel-Level Labeling with Gradient-Based Inference