AI Paper Atlas

Research map / Video / Video segmentation and tracking

STDDNet: Harnessing Mamba for Video Polyp Segmentation via Spatial-aligned Temporal Modeling and Discriminative Dynamic Representation Learning

Guilian Chen, Huisi Wu, Jing Qin

ICCV 2025 · Video › Video segmentation and tracking · approach: Video semantic and panoptic segmentation · 6 citations (OpenAlex)

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 changedWhereMeasuredEffect
add STDP module (ablation)PVTv2-B2 baseline · SUN-SEG-EasyDice (%): 82.73 -> 84.73Improved
add STDP module (ablation)PVTv2-B2 baseline · SUN-SEG-HardDice (%): 80.29 -> 83.48Improved
add DDFE module (ablation)PVTv2-B2 baseline · SUN-SEG-EasyDice (%): 82.73 -> 84.54Improved
add DDFE module (ablation)PVTv2-B2 baseline · SUN-SEG-HardDice (%): 80.29 -> 83.79Improved
add DDFE on top of STDP (ablation)STDDNet · SUN-SEG-EasyDice (%): 84.73 -> 85.86Improved
add random masking in STDP (ablation)STDDNet · SUN-SEG-HardDice (%): 83.06 -> 83.48Improved
add filter in STDP (ablation)STDDNet · SUN-SEG-HardDice (%): 83.23 -> 83.48Improved
add mask guidance in STDP (ablation)STDDNet · SUN-SEG-HardDice (%): 82.95 -> 83.48Improved
add position compensation in DDFE (ablation)STDDNet · SUN-SEG-HardDice (%): 82.16 -> 83.79Improved
add reverse operation in DDFE (ablation)STDDNet · SUN-SEG-HardDice (%): 83.15 -> 83.79Improved
add activation ratio in DDFE (ablation)STDDNet · SUN-SEG-HardDice (%): 82.82 -> 83.79Improved
use PVTv2-B2 instead of Res2Net-50 backboneSTDDNet · SUN-SEG-HardDice (%): 83.45 -> 84.59Improved

Later papers that compared with this method

1 later accepted paper reports STDDNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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.

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

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy