Generalized background subtraction using superpixels with label integrated motion estimation
- Authors
- Lim, Jongwoo; Han, Bohyung
- Issue Date
- Sep-2014
- Publisher
- Springer Verlag
- Keywords
- density propagation; generalized background subtraction; layered optical flow estimation; superpixel segmentation
- Citation
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), v.8693 LNCS, no.PART 5, pp.173 - 187
- Indexed
- SCOPUS
- Journal Title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- Volume
- 8693 LNCS
- Number
- PART 5
- Start Page
- 173
- End Page
- 187
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/159162
- DOI
- 10.1007/978-3-319-10602-1_12
- ISSN
- 0302-9743
- Abstract
- We propose an online background subtraction algorithm with superpixel-based density estimation for videos captured by moving camera. Our algorithm maintains appearance and motion models of foreground and background for each superpixel, computes foreground and background likelihoods for each pixel based on the models, and determines pixelwise labels using binary belief propagation. The estimated labels trigger the update of appearance and motion models, and the above steps are performed iteratively in each frame. After convergence, appearance models are propagated through a sequential Bayesian filtering, where predictions rely on motion fields of both labels whose computation exploits the segmentation mask. Superpixel-based modeling and label integrated motion estimation make propagated appearance models more accurate compared to existing methods since the models are constructed on visually coherent regions and the quality of estimated motion is improved by avoiding motion smoothing across regions with different labels. We evaluate our algorithm with challenging video sequences and present significant performance improvement over the state-of-the-art techniques quantitatively and qualitatively.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 컴퓨터소프트웨어학부 > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/159162)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.