Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Video Object Segmentation using Space-Time Memory Networks

About

We propose a novel solution for semi-supervised video object segmentation. By the nature of the problem, available cues (e.g. video frame(s) with object masks) become richer with the intermediate predictions. However, the existing methods are unable to fully exploit this rich source of information. We resolve the issue by leveraging memory networks and learn to read relevant information from all available sources. In our framework, the past frames with object masks form an external memory, and the current frame as the query is segmented using the mask information in the memory. Specifically, the query and the memory are densely matched in the feature space, covering all the space-time pixel locations in a feed-forward fashion. Contrast to the previous approaches, the abundant use of the guidance information allows us to better handle the challenges such as appearance changes and occlussions. We validate our method on the latest benchmark sets and achieved the state-of-the-art performance (overall score of 79.4 on Youtube-VOS val set, J of 88.7 and 79.2 on DAVIS 2016/2017 val set respectively) while having a fast runtime (0.16 second/frame on DAVIS 2016 val set).

Seoung Wug Oh, Joon-Young Lee, Ning Xu, Seon Joo Kim• 2019

Related benchmarks

TaskDatasetResultRank
Video Object SegmentationDAVIS 2017 (val)
J mean81.7
1130
Video Object SegmentationDAVIS 2016 (val)
J Mean88.7
564
Video Object SegmentationYouTube-VOS 2018 (val)
J Score (Seen)79.7
493
Action RecognitionSomething-Something v2 (test)
Top-1 Acc64.2
333
Video Object SegmentationDAVIS 2017 (test-dev)
Region J Mean69.3
237
Video Object SegmentationYouTube-VOS 2019 (val)
J-Score (Seen)79.8
231
Visual Object TrackingVOT 2020 (test)
EAO0.308
147
Video Object SegmentationDAVIS 2017 (test)
J (Jaccard Index)79.2
107
Video Object SegmentationYouTube-VOS (val)
J Score (Seen)79.7
81
Video Object SegmentationYouTube-VOS 2018
Score G79.4
47
Showing 10 of 34 rows

Other info

Follow for update