Top-Down Learning for Structured Labeling with Convolutional Pseudoprior

Abstract

Current practice in convolutional neural networks (CNN) remains largely bottom-up and the role of top-down process in CNN for pattern analysis and visual inference is not very clear. In this paper, we propose a new method for structured labeling by developing convolutional pseudo-prior (ConvPP) on the ground-truth labels. Our method has several interesting properties: (1) compared with classical machine learning algorithms like CRFs and Structural SVM, ConvPP automatically learns rich convolutional kernels to capture both short- and long- range contexts; (2) compared with cascade classifiers like Auto-Context, ConvPP avoids the iterative steps of learning a series of discriminative classifiers and automatically learns contextual configurations; (3) compared with recent efforts combing CNN models with CRFs and RNNs, ConvPP learns convolution in the labeling space with much improved modeling capability and less manual specification; (4) compared with Bayesian models like MRFs, ConvPP capitalizes on the rich representation power of convolution by automatically learning priors built on convolutional filters. We accomplish our task using pseudo-likelihood approximation to the prior under a novel fixed-point network structure that facilitates an end-to-end learning process. We show state-of-the-art results on sequential labeling and image labeling benchmarks.

To appear in ECCV 2016, 16 pages, 6 figures

References in corpus (6)

• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• Conditional Random Fields as Recurrent Neural Networks
• Deep Convolutional Networks on Graph-Structured Data
• Convolutional Feature Masking for Joint Object and Stuff Segmentation
• Winner-Take-All Autoencoders
• BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation

Cited by in corpus (9)

• PixelNet: Representation of the pixels, by the pixels, and for the pixels
• PixelNet: Towards a General Pixel-level Architecture
• Iterative Visual Reasoning Beyond Convolutions
• Improving Fully Convolution Network for Semantic Segmentation
• ORDNet: Capturing Omni-Range Dependencies for Scene Parsing
• Dense Recurrent Neural Networks for Scene Labeling
• Neuron-level Selective Context Aggregation for Scene Segmentation
• Spatial Memory for Context Reasoning in Object Detection
• Scene Parsing via Dense Recurrent Neural Networks with Attentional Selection