Occlusion Boundary Detection via Deep Exploration of Context

Huan Fu (1) , Chaohui Wang (2) , Dacheng Tao (1) , Michael J Black (3)

1 QCIS and FEIT
2 LIGM - Laboratoire d'Informatique Gaspard-Monge
3 Max Planck Institute for Intelligent Systems [Tübingen]

Occlusion boundaries contain rich perceptual information about the underlying scene structure. They also provide important cues in many visual perception tasks such as scene understanding, object recognition, and segmentation. In this paper, we improve occlusion boundary detection via enhanced exploration of contextual information (e.g., local structural boundary patterns, observations from surrounding regions, and temporal context), and in doing so develop a novel approach based on convolutional neural networks (CNNs) and conditional random fields (CRFs). Experimental results demonstrate that our detector significantly out-performs the state-of-the-art (e.g., improving the F-measure from 0.62 to 0.71 on the commonly used CMU benchmark). Last but not least, we empirically assess the roles of several important components of the proposed detector, so as to validate the rationale behind this approach.

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Huan Fu, Chaohui Wang, Dacheng Tao, Michael J Black. Occlusion Boundary Detection via Deep Exploration of Context. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun 2016, Las Vegas, United States. pp.241 - 250, ⟨10.1109/CVPR.2016.33⟩. ⟨hal-01578439⟩