1. Abstract

Pixel-level labelling tasks, such as semantic segmentation, play a central role in image understanding.

[info]
labelling task：标签任务
semantic segmentation：语义切分

Recent approaches have attempted to harness the capabilities of deep learning techniques for image recognition to tackle pixel level labelling tasks.

[info] harness：利用

One central issue in this methodology is the limited capacity of deep learning techniques to delineate visual objects.

[info] delineate：描绘

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.

[success] CNN + CRF

To this end, we formulate Conditional Random Fields with Gaussian pairwise potentials and mean-field approximate inference as Recurrent Neural Networks.

[warning]
to this end：为了这个目的
[?] Gaussian pairwise potentials and mean-field approximate inference

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.