Papers › Video Object Segmentation with Adaptive Feature Bank and Uncertain-Region Refinement

Video Object Segmentation with Adaptive Feature Bank and Uncertain-Region Refinement

15 Oct 2020NeurIPS 2020 12arXiv:2010.07958archive 2025-07-28

Yongqing Liang, Xin Li, Navid Jafari, Qin Chen

We propose a new matching-based framework for semi-supervised video object segmentation (VOS). Recently, state-of-the-art VOS performance has been achieved by matching-based algorithms, in which feature banks are created to store features for region matching and classification. However, how to effectively organize information in the continuously growing feature bank remains under-explored, and this leads to inefficient design of the bank. We introduce an adaptive feature bank update scheme to dynamically absorb new features and discard obsolete features. We also design a new confidence loss and a fine-grained segmentation module to enhance the segmentation accuracy in uncertain regions. On public benchmarks, our algorithm outperforms existing state-of-the-arts.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

xmlyqing00/AFB-URR officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

SegmentationSemantic SegmentationSemi-Supervised Video Object SegmentationVideo Object SegmentationVideo Semantic Segmentation

Datasets

Introduced by this paper, per the archive.

Long Video DatasetLong Video Dataset (3X)

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semi-Supervised Video Object Segmentation DAVIS (no YouTube-VOS training) AFB-URR D17 val (F) 76.1 #12 of 26 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS (no YouTube-VOS training) AFB-URR D17 val (G) 74.6 #12 of 26 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS (no YouTube-VOS training) AFB-URR D17 val (J) 73.0 #12 of 26 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS (no YouTube-VOS training) AFB-URR FPS 4.00 #12 of 26 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR F-measure (Decay) 15.5 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR F-measure (Mean) 76.1 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR F-measure (Recall) 87.0 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR J&F 74.6 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR Jaccard (Decay) 13.8 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR Jaccard (Mean) 73.0 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation DAVIS 2017 (val) AFB-URR Jaccard (Recall) 85.3 #56 of 81 Archive leaderboard report
Semi-Supervised Video Object Segmentation Long Video Dataset AFB-URR F 84.5 #6 of 9 Archive leaderboard report
Semi-Supervised Video Object Segmentation Long Video Dataset AFB-URR J 82.9 #6 of 9 Archive leaderboard report
Semi-Supervised Video Object Segmentation Long Video Dataset AFB-URR J&F 83.7 #6 of 9 Archive leaderboard report
Semi-Supervised Video Object Segmentation Long Video Dataset (3X) AFB-URR F 84.6 #2 of 2 Archive leaderboard report
Semi-Supervised Video Object Segmentation Long Video Dataset (3X) AFB-URR J 82.9 #2 of 2 Archive leaderboard report
Semi-Supervised Video Object Segmentation Long Video Dataset (3X) AFB-URR J&F 83.8 #2 of 2 Archive leaderboard report
Semi-Supervised Video Object Segmentation YouTube-VOS 2018 AFB-URR F-Measure (Seen) 83.1 #41 of 53 Archive leaderboard report
Semi-Supervised Video Object Segmentation YouTube-VOS 2018 AFB-URR F-Measure (Unseen) 82.6 #41 of 53 Archive leaderboard report
Semi-Supervised Video Object Segmentation YouTube-VOS 2018 AFB-URR Jaccard (Seen) 78.8 #41 of 53 Archive leaderboard report
Semi-Supervised Video Object Segmentation YouTube-VOS 2018 AFB-URR Jaccard (Unseen) 74.1 #41 of 53 Archive leaderboard report
Semi-Supervised Video Object Segmentation YouTube-VOS 2018 AFB-URR Overall 79.6 #41 of 53 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

VOS

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections