Detect, Replace, Refine: Deep Structured Prediction For Pixel Wise Labeling

TL;DR

This paper introduces a novel deep structured prediction architecture for pixel-wise labeling that decomposes the task into detection, replacement, and refinement steps, achieving state-of-the-art results in disparity estimation.

Contribution

It proposes a generic architecture that explicitly models label correction steps, outperforming prior direct prediction methods in dense disparity estimation.

Findings

Achieves state-of-the-art disparity estimation on KITTI 2015

Outperforms previous methods significantly in quantitative metrics

Provides extensive evaluation of different architecture variants

Abstract

Pixel wise image labeling is an interesting and challenging problem with great significance in the computer vision community. In order for a dense labeling algorithm to be able to achieve accurate and precise results, it has to consider the dependencies that exist in the joint space of both the input and the output variables. An implicit approach for modeling those dependencies is by training a deep neural network that, given as input an initial estimate of the output labels and the input image, it will be able to predict a new refined estimate for the labels. In this context, our work is concerned with what is the optimal architecture for performing the label improvement task. We argue that the prior approaches of either directly predicting new label estimates or predicting residual corrections w.r.t. the initial labels with feed-forward deep network architectures are sub-optimal. Instead, we propose a generic architecture that decomposes the label improvement task to three steps: 1) detecting the initial label estimates that are incorrect, 2) replacing the incorrect labels with new ones, and finally 3) refining the renewed labels by predicting residual corrections w.r.t. them. Furthermore, we explore and compare various other alternative architectures that consist of the aforementioned Detection, Replace, and Refine components. We extensively evaluate the examined architectures in the challenging task of dense disparity estimation (stereo matching) and we report both quantitative and qualitative results on three different datasets. Finally, our dense disparity estimation network that implements the proposed generic architecture, achieves state-of-the-art results in the KITTI 2015 test surpassing prior approaches by a significant margin.