Human instance segmentation from video using detector-based conditional random fields

Conference item

In this work, we propose a method for instance based human segmentation in images and videos, extending the recent detector-based conditional random field model of Ladicky et.al. Instance based human segmentation involves pixel level labeling of an image, partitioning it into distinct human instances and background. To achieve our goal, we add three new components to their framework. First, we include human parts-based detection potentials to take advantage of the structure present in human instances. Further, in order to generate a consistent segmentation from different human parts, we incorporate shape prior information, which biases the segmentation to characteristic overall human shapes. Also, we enhance the representative power of the energy function by adopting exemplar instance based matching terms, which helps our method to adapt easily to different human sizes and poses. Finally, we extensively evaluate our proposed method on the Buffy dataset with our new segmented ground truth images, and show a substantial improvement over existing CRF methods. These new annotations will be made available for future use as well. © 2011. The copyright of this document resides with its authors.

Authors: Vineet, V; Warrell, J; Ladický, L; Torr, PHS

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: British Machine Vision Association
• Host title: Proceedings of the British Machine Vision Conference 2011
• Publication date: 2011-01-01
• Event title: British Machine Vision Conference (BMVC 2011)
• Event location: Dundee
• Event website: https://bmva-archive.org.uk/bmvc/2011/proceedings/frontmatter.html
• Event start date: 2011-08-29
• Event end date: 2011-09-02
• ISBN: 190172543X
• Language: English