CVFeb 11, 2015

Conditional Random Fields as Recurrent Neural Networks

arXiv:1502.03240v32586 citations
AI Analysis

This work addresses the limited object delineation capabilities in deep learning for image segmentation, offering a novel integration that improves accuracy for computer vision applications.

The paper tackles the problem of pixel-level labeling in semantic segmentation by integrating Conditional Random Fields (CRFs) with Convolutional Neural Networks (CNNs) into a single end-to-end trainable network, achieving top results on the Pascal VOC 2012 benchmark.

Pixel-level labelling tasks, such as semantic segmentation, play a central role in image understanding. Recent approaches have attempted to harness the capabilities of deep learning techniques for image recognition to tackle pixel-level labelling tasks. One central issue in this methodology is the limited capacity of deep learning techniques to delineate visual objects. To solve this problem, we introduce a new form of convolutional neural network that combines the strengths of Convolutional Neural Networks (CNNs) and Conditional Random Fields (CRFs)-based probabilistic graphical modelling. To this end, we formulate mean-field approximate inference for the Conditional Random Fields with Gaussian pairwise potentials as Recurrent Neural Networks. This network, called CRF-RNN, is then plugged in as a part of a CNN to obtain a deep network that has desirable properties of both CNNs and CRFs. Importantly, our system fully integrates CRF modelling with CNNs, making it possible to train the whole deep network end-to-end with the usual back-propagation algorithm, avoiding offline post-processing methods for object delineation. We apply the proposed method to the problem of semantic image segmentation, obtaining top results on the challenging Pascal VOC 2012 segmentation benchmark.

Code Implementations6 repos
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes