Abstract

We focus on the problem of semantic segmentation based on RGB-D data, with emphasis on analyzing cluttered indoor scenes containing many visual categories and instances. Our approach is based on a parametric fifigureground intensity and depth-constrained proposal process that generates spatial layout hypotheses at multiple locations and scales in the image followed by a sequential inference algorithm that produces a complete scene estimate. Our contributions can be summarized as follows: (1) a generalization of parametric max flflow fifigure-ground proposal methodology to take advantage of intensity and depth information, in order to systematically and effificiently generate the breakpoints of an underlying spatial model in polynomial time, (2) new region description methods based on second-order pooling over multiple features constructed using both intensity and depth channels, (3) a principled search-based structured prediction inference and learning process that resolves conflflicts in overlapping spatial partitions and selects regions sequentially towards complete scene estimates, and (4) extensive evaluation of the impact of depth, as well as the effectiveness of a large number of descriptors, both pre-designed and automatically obtained using deep learning, in a diffificult RGB-D semantic segmentation problem with 92 classes. We report state of the art results in the challenging NYU Depth Dataset V2 [44], extended for the RMRC 2013 and RMRC 2014 Indoor Segmentation Challenges, where currently the proposed model ranks fifirst. Moreover, we show that by combining secondorder and deep learning features, over 15% relative accuracy improvements can be additionally achieved. In a scene classifification benchmark, our methodology further improves the state of the art by 24%