Image Parsing with a Wide Range of Classes and Scene-Level Context

Abstract

This paper presents a nonparametric scene parsing approach that improves the overall accuracy, as well as the coverage of foreground classes in scene images. We first improve the label likelihood estimates at superpixels by merging likelihood scores from different probabilistic classifiers. This boosts the classification performance and enriches the representation of less-represented classes. Our second contribution consists of incorporating semantic context in the parsing process through global label costs. Our method does not rely on image retrieval sets but rather assigns a global likelihood estimate to each label, which is plugged into the overall energy function. We evaluate our system on two large-scale datasets, SIFTflow and LMSun. We achieve state-of-the-art performance on the SIFTflow dataset and near-record results on LMSun.

Published at CVPR 2015, Computer Vision and Pattern Recognition (CVPR), 2015 IEEE Conference on

References in corpus (1)

• Scene Parsing with Multiscale Feature Learning, Purity Trees, and Optimal Covers

Cited by in corpus (7)

• Scene Parsing with Integration of Parametric and Non-parametric Models
• Multi-hypothesis contextual modeling for semantic segmentation
• Learning Deep Representations for Scene Labeling with Semantic Context Guided Supervision
• Removing Class Imbalance using Polarity-GAN: An Uncertainty Sampling Approach
• Sample and Filter: Nonparametric Scene Parsing via Efficient Filtering
• Scene Labeling Through Knowledge-Based Rules Employing Constrained Integer Linear Programing
• Region-based semantic segmentation with end-to-end training