Deep GrabCut for Object Selection

Ning Xu, Brian Price, Scott Cohen, Jimei Yang and Thomas Huang

Abstract

Most previous bounding-box-based segmentation methods assume the bounding box tightly covers the object of interest. However it is common that a rectangle input could be too large or too small. In this paper, we propose a novel segmentation approach that uses a rectangle as a soft constraint by transforming it into an Euclidean distance map. A convolutional encoder-decoder network is trained end-to-end by concatenating images with these distance maps as inputs and predicting the object masks as outputs. Our approach gets accurate segmentation results given sloppy rectangles while being general for both interactive segmentation and instance segmentation. We show our network extends to curve-based input without retraining. We further apply our network to instance-level semantic segmentation and resolve any overlap using a conditional random field.

Session

Posters

Files

Paper (PDF)
Supplementary (PDF)

DOI

10.5244/C.31.182
https://dx.doi.org/10.5244/C.31.182

Citation

Bibtex

            @inproceedings{BMVC2017_182,
                title={Deep GrabCut for Object Selection},
                author={Ning Xu, Brian Price, Scott Cohen, Jimei Yang and Thomas Huang},
                year={2017},
                month={September},
                pages={182.1-182.12},
                articleno={182},
                numpages={12},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.182},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.182}
            }