SALI: Short-term Alignment and Long-term Interaction Network for Colonoscopy Video Polyp Segmentation

TL;DR

The SALI network enhances colonoscopy video polyp segmentation by combining short-term spatial alignment and long-term memory modules, improving robustness and accuracy in challenging low-quality frames.

Contribution

This paper introduces the SALI network, integrating SAM and LIM modules for improved robustness in video polyp segmentation, addressing spatial incoherence and low-quality frame challenges.

Findings

Outperforms current state-of-the-art methods on SUNSEG benchmark.

Improves Dice coefficient by up to 4.1% on test subsets.

Demonstrates robustness to spatial variations and low-visual cues.

Abstract

Colonoscopy videos provide richer information in polyp segmentation for rectal cancer diagnosis. However, the endoscope's fast moving and close-up observing make the current methods suffer from large spatial incoherence and continuous low-quality frames, and thus yield limited segmentation accuracy. In this context, we focus on robust video polyp segmentation by enhancing the adjacent feature consistency and rebuilding the reliable polyp representation. To achieve this goal, we in this paper propose SALI network, a hybrid of Short-term Alignment Module (SAM) and Long-term Interaction Module (LIM). The SAM learns spatial-aligned features of adjacent frames via deformable convolution and further harmonizes them to capture more stable short-term polyp representation. In case of low-quality frames, the LIM stores the historical polyp representations as a long-term memory bank, and explores the retrospective relations to interactively rebuild more reliable polyp features for the current segmentation. Combing SAM and LIM, the SALI network of video segmentation shows a great robustness to the spatial variations and low-visual cues. Benchmark on the large-scale SUNSEG verifies the superiority of SALI over the current state-of-the-arts by improving Dice by 2.1%, 2.5%, 4.1% and 1.9%, for the four test sub-sets, respectively. Codes are at https://github.com/Scatteredrain/SALI.