Research map / Video / Video segmentation and tracking
STEm-Seg: Spatio-temporal Embeddings for Instance Segmentation in Videos
Ali Athar, Sabarinath Mahadevan, Aljosa Osep, Laura Leal-Taixé, Bastian Leibe
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add spatial coordinate offset to the embedding mixing function (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 57.3 -> 61.6 | Improved |
| add a temporal coordinate offset dimension (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 61.6 -> 62.6 | Improved |
| use a free embedding dimension instead of a temporal dimension (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 62.6 -> 62.8 | Improved |
| use E=4 with two free dimensions instead of temporal+free (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 64.2 -> 64.4 | Improved |
| add a fifth embedding dimension (E=5) (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 64.4 -> 62.4 | Hurt |
| increase clip length from T=4 to T=8 (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 62.2 -> 64.4 | Improved |
| increase clip length from T=8 to T=16 (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 64.4 -> 64.7 | Improved |
| increase clip length from T=16 to T=24 (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 64.7 -> 63.1 | Hurt |
| add video data to image-only training (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 57.1 -> 64.4 | Improved |
| add image-augmented data to video-only training (ablation) | STEm-Seg · DAVIS 2019 unsupervised val | J&F: 60.7 -> 64.4 | Improved |
| use oracle category labels instead of predicted ones (ablation) | STEm-Seg · YouTube-VIS custom val split | AP: 24.7 -> 33.0 | Improved |
| use ResNet-101 instead of ResNet-50 backbone | STEm-Seg · YouTube-VIS val | AP: 30.6 -> 34.6 | Improved |
Later papers that compared with this method
11 later accepted papers report STEm-Seg as a baseline in their results tables; 11 of them report a better number on most of the results they share with it.
- Exposing the Self-Supervised Space-Time Correspondence Learning via Graph Kernels (AAAI 2023 · better on 1 of 1 shared result)
- Occluded Video Instance Segmentation: A Benchmark (IJCV 2022 · better on 2 of 2 shared results)
- SeqFormer: Sequential Transformer for Video Instance Segmentation (ECCV 2022 · better on 1 of 1 shared result)
- End-to-End Video Instance Segmentation via Spatial-Temporal Graph Neural Networks (ICCV 2021 · better on 3 of 3 shared results)
- Video Instance Segmentation With a Propose-Reduce Paradigm (ICCV 2021 · better on 3 of 3 shared results)
- End-to-End Video Instance Segmentation With Transformers (CVPR 2021 · better on 2 of 2 shared results)
- Spatial Feature Calibration and Temporal Fusion for Effective One-Stage Video Instance Segmentation (CVPR 2021 · better on 2 of 2 shared results)
- Occluded Video Instance Segmentation: Dataset and ICCV 2021 Challenge (NeurIPS 2021 · better on 2 of 2 shared results)
3 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FEELVOS — FEELVOS: Fast End-To-End Embedding Learning for Video Object Segmentation (CVPR 2019) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Less than Few: Self-Shot Video Instance Segmentation (ECCV 2022 · similarity 0.89)
- InsPro: Propagating Instance Query and Proposal for Online Video Instance Segmentation (NeurIPS 2022 · similarity 0.89)
- Hybrid Instance-Aware Temporal Fusion for Online Video Instance Segmentation (AAAI 2022 · similarity 0.88)
- SyncVIS: Synchronized Video Instance Segmentation (NeurIPS 2024 · similarity 0.88)
- Know Where You Are From: Event-Based Segmentation via Spatio-Temporal Propagation (AAAI 2025 · similarity 0.87)
- Learning Spatial-Semantic Features for Robust Video Object Segmentation (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
