CASIA OpenIR  > 模式识别国家重点实验室  > 先进数据分析与学习
Greedy Batch-Based Minimum-Cost Flows for Tracking Multiple Objects
Wang, Xinchao1; Fan, Bin2; Chang, Shiyu3; Wang, Zhangyang4; Liu, Xianming1,5; Tao, Dacheng6,7; Huang, Thomas S.1; Bin Fan
Source PublicationIEEE TRANSACTIONS ON IMAGE PROCESSING
2017-10-01
Volume26Issue:10Pages:4765-4776
SubtypeArticle
AbstractMinimum-cost flow algorithms have recently achieved state-of-the-art results in multi-object tracking. However, they rely on the whole image sequence as input. When deployed in real-time applications or in distributed settings, these algorithms first operate on short batches of frames and then stitch the results into full trajectories. This decoupled strategy is prone to errors because the batch-based tracking errors may propagate to the final trajectories and cannot be corrected by other batches. In this paper, we propose a greedy batch-based minimum-cost flow approach for tracking multiple objects. Unlike existing approaches that conduct batch-based tracking and stitching sequentially, we optimize consecutive batches jointly so that the tracking results on one batch may benefit the results on the other. Specifically, we apply a generalized minimum-cost flows (MCF) algorithm on each batch and generate a set of conflicting trajectories. These trajectories comprise the ones with high probabilities, but also those with low probabilities potentially missed by detectors and trackers. We then apply the generalized MCF again to obtain the optimal matching between trajectories from consecutive batches. Our proposed approach is simple, effective, and does not require training. We demonstrate the power of our approach on data sets of different scenarios.
KeywordMulti-object Tracking Minimum-cost Flows Batch Processing Graph Transformation
WOS HeadingsScience & Technology ; Technology
DOI10.1109/TIP.2017.2723239
WOS KeywordMULTITARGET TRACKING ; PROPAGATION ; MODELS
Indexed BySCI
Language英语
Funding OrganizationSwiss National Science Foundation ; Natural Science Foundation of China(61573352 ; Australian Research Council(FT-130101457 ; 61403375 ; DP-140102164 ; 61472119) ; LP-150100671)
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000406329500014
Citation statistics
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/19694
Collection模式识别国家重点实验室_先进数据分析与学习
Corresponding AuthorBin Fan
Affiliation1.Univ Illinois, Beckman Inst, Image Format & Proc Grp, Urbana, IL 61801 USA
2.Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China
3.IBM Thomas J Watson Res Ctr, Yorktown Hts, NY 10598 USA
4.Texas A&M Univ, Dept Comp Sci & Engn, College Stn, TX 77843 USA
5.Facebook Inc, San Francisco, CA 94025 USA
6.UBTech Sydney Artificial Intelligence Inst, Sydney, NSW 2008, Australia
7.Univ Sydney, Fac Engn & Informat Technol, Sch Informat Technol, Sydney, NSW 2008, Australia
Recommended Citation
GB/T 7714
Wang, Xinchao,Fan, Bin,Chang, Shiyu,et al. Greedy Batch-Based Minimum-Cost Flows for Tracking Multiple Objects[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2017,26(10):4765-4776.
APA Wang, Xinchao.,Fan, Bin.,Chang, Shiyu.,Wang, Zhangyang.,Liu, Xianming.,...&Bin Fan.(2017).Greedy Batch-Based Minimum-Cost Flows for Tracking Multiple Objects.IEEE TRANSACTIONS ON IMAGE PROCESSING,26(10),4765-4776.
MLA Wang, Xinchao,et al."Greedy Batch-Based Minimum-Cost Flows for Tracking Multiple Objects".IEEE TRANSACTIONS ON IMAGE PROCESSING 26.10(2017):4765-4776.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wang, Xinchao]'s Articles
[Fan, Bin]'s Articles
[Chang, Shiyu]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wang, Xinchao]'s Articles
[Fan, Bin]'s Articles
[Chang, Shiyu]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wang, Xinchao]'s Articles
[Fan, Bin]'s Articles
[Chang, Shiyu]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.