Research map / Video / Video segmentation and tracking
GDKVM: Echocardiography Video Segmentation via Spatiotemporal Key-Value Memory with Gated Delta Rule
Rui Wang, Yimu Sun, Jingxing Guo, Huisi Wu, Jing Qin
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 Gated Delta Rule to LKVA (ablation) | GDKVM · CAMUS | mDice: 93.10 -> 94.49 | Improved |
| add Key-Pixel Feature Fusion to LKVA (ablation) | GDKVM · CAMUS | mDice: 93.10 -> 93.30 | Improved |
| add KPFF to LKVA+GDR (ablation) | GDKVM · CAMUS | mDice: 94.49 -> 95.11 | Improved |
| add GDR to LKVA+KPFF (ablation) | GDKVM · CAMUS | mDice: 93.30 -> 95.11 | Improved |
| add decay alpha_t to the delta-rule update (ablation) | GDR · CAMUS | mDice: 94.57 -> 95.11 | Improved |
| add write strength beta_t to the decay update (ablation) | GDR · CAMUS | mDice: 94.26 -> 95.11 | Improved |
| use full GDR update instead of additive update (ablation) | GDKVM · CAMUS | mDice: 93.30 -> 95.11 | Improved |
| directly replace old memory with new information (no gates) (ablation) | GDKVM · CAMUS | mDice: 93.30 -> 74.68 | Hurt |
| use full GDR instead of additive update (ablation) | GDKVM · CAMUS | inference time (ms): 151.61 -> 160.62 | Hurt |
Later papers that compared with this method
1 later accepted paper reports GDKVM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- OSA: Echocardiography Video Segmentation via Orthogonalized State Update and Anatomical Prior-aware Feature Enhancement (CVPR 2026 · better on 5 of 6 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- MemSAM — MemSAM: Taming Segment Anything Model for Echocardiography Video Segmentation (CVPR 2024) · better on 3 of 3 shared results
- VideoMamba — VideoMamba: State Space Model for Efficient Video Understanding (ECCV 2024) · better on 3 of 3 shared results
- XMem++ — XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model (ECCV 2022) · better on 3 of 3 shared results
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
- OSA: Echocardiography Video Segmentation via Orthogonalized State Update and Anatomical Prior-aware Feature Enhancement (CVPR 2026 · similarity 0.89)
- MemSAM: Taming Segment Anything Model for Echocardiography Video Segmentation (CVPR 2024 · similarity 0.86)
- EchoONE: Segmenting Multiple Echocardiography Planes in One Model (CVPR 2025 · similarity 0.85)
- Self-supervised Learning of Echocardiographic Video Representations via Online Cluster Distillation (NeurIPS 2025 · similarity 0.84)
- EchoXFlow: A Beamspace Echocardiography Dataset for Cardiac Motion, Flow, and Function (NeurIPS 2026 · similarity 0.83)
- Bidirectional Recurrence for Cardiac Motion Tracking with Gaussian Process Latent Coding (NeurIPS 2024 · similarity 0.83)
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.
