Improved Image Boundaries for Better Video Segmentation

Anna Khoreva, Rodrigo Benenson, Fabio Galasso,

Matthias Hein and Bernt Schiele

Abstract

Graph-based video segmentation methods rely on superpixels as starting point. While most previous work has focused on the construction of the graph edges and weights as well as solving the graph partitioning problem, this paper focuses on better superpixels for video segmentation. We demonstrate by a comparative analysis that superpixels extracted from boundaries perform best, and show that
boundary estimation can be significantly improved via image and time domain cues. With superpixels generated from our better boundaries we observe consistent improvement for two video segmentation methods in two different datasets.


Figure 1. Graph based video segmentation relies on having high quality superpixels (voxels) as starting point (graph nodes). We explore diverse techniques to improve boundary estimates, which result in better superpixels, which in turn has a significant impact on final video segmentation.

Results


Figure 2. Comparison of video segmentation results to one human ground truth. Our superpixels allow for the video segmentation methods to better distinguish visual objects and to limit label leakage due to inherent temporal smoothness of the boundaries. The last row shows a failure case for all methods.
   
   

Data

ArXiv  Paper   Presentation slides

For further information or data, please contact Anna Khoreva <khoreva at mpi-inf.mpg.de>.

References

[Khoreva et al., 2016] Improved Image Boundaries for Better Video Segmentation, A. Khoreva, R. Benenson, F. Galasso,  M. Hein and B. Schiele, ECCV Workshops, (2016)

@inproceedings{khoreva_ECCV16,
title={Improved Image Boundaries for Better Video Segmentation
},
author={A. Khoreva and R. Benenson and F. Galasso and M. Hein and B. Schiele},
booktitle={
European Conference on Computer Vision Workshops},
year={2016}}