Papers › STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos

STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos

18 Mar 2020ECCV 2020 8arXiv:2003.08429archive 2025-07-28

Ali Athar, Sabarinath Mahadevan, Aljoša Ošep, Laura Leal-Taixé, Bastian Leibe

Existing methods for instance segmentation in videos typically involve multi-stage pipelines that follow the tracking-by-detection paradigm and model a video clip as a sequence of images. Multiple networks are used to detect objects in individual frames, and then associate these detections over time. Hence, these methods are often non-end-to-end trainable and highly tailored to specific tasks. In this paper, we propose a different approach that is well-suited to a variety of tasks involving instance segmentation in videos. In particular, we model a video clip as a single 3D spatio-temporal volume, and propose a novel approach that segments and tracks instances across space and time in a single stage. Our problem formulation is centered around the idea of spatio-temporal embeddings which are trained to cluster pixels belonging to a specific object instance over an entire video clip. To this end, we introduce (i) novel mixing functions that enhance the feature representation of spatio-temporal embeddings, and (ii) a single-stage, proposal-free network that can reason about temporal context. Our network is trained end-to-end to learn spatio-temporal embeddings as well as parameters required to cluster these embeddings, thus simplifying inference. Our method achieves state-of-the-art results across multiple datasets and tasks. Code and models are available at https://github.com/sabarim/STEm-Seg.

PaperPDFConference PDFCode

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

Code

sabarim/STEm-Seg 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

Instance SegmentationSemantic SegmentationUnsupervised Video Object SegmentationVideo Instance Segmentation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Unsupervised Video Object Segmentation DAVIS 2017 (val) STEm-Seg F-measure (Mean) 67.8 #5 of 10 Archive leaderboard report
Unsupervised Video Object Segmentation DAVIS 2017 (val) STEm-Seg F-measure (Recall) 75.5 #5 of 10 Archive leaderboard report
Unsupervised Video Object Segmentation DAVIS 2017 (val) STEm-Seg J&F 64.7 #5 of 10 Archive leaderboard report
Unsupervised Video Object Segmentation DAVIS 2017 (val) STEm-Seg Jaccard (Mean) 61.5 #5 of 10 Archive leaderboard report
Unsupervised Video Object Segmentation DAVIS 2017 (val) STEm-Seg Jaccard (Recall) 70.4 #5 of 10 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-101) AP50 55.8 #35 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-101) AP75 37.9 #35 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-101) AR1 34.4 #35 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-101) AR10 41.6 #35 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-101) mask AP 34.6 #35 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-50) AP50 50.7 #40 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-50) AP75 37.9 #40 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-50) AR1 34.4 #40 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-50) AR10 41.6 #40 of 44 Archive leaderboard report
Video Instance Segmentation YouTube-VIS validation STEm-Seg (ResNet-50) mask AP 30.6 #40 of 44 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.

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