Yifei Zhao

Accurate and efficient Video Polyp Segmentation (VPS) is vital for the early detection of colorectal cancer and the effective treatment of polyps. However, achieving this remains highly challenging due to the inherent difficulty in modeling the spatial-temporal relationships within colonoscopy videos. Existing methods that directly associate video frames frequently fail to account for variations in polyp or background motion, leading to excessive noise and reduced segmentation accuracy. Conversely, approaches that rely on optical flow models to estimate motion and align frames incur significant computational overhead. To address these limitations, we propose a novel VPS framework, termed Deformable Alignment and Local Attention (DALA). In this framework, we first construct a shared encoder to jointly encode the feature representations of paired video frames. Subsequently, we introduce a Multi-Scale Frame Alignment (MSFA) module based on deformable convolution to estimate the motion between reference and anchor frames. The multi-scale architecture is designed to accommodate the scale variations of polyps arising from differing viewing angles and speeds during colonoscopy. Furthermore, Local Attention (LA) is employed to selectively aggregate the aligned features, yielding more precise spatial-temporal feature representations. Extensive experiments conducted on the challenging SUN-SEG dataset and PolypGen dataset demonstrate that DALA achieves superior performance compared to stateof-the-art models. The code will be publicly available at https://github.com/xff12138/DALA.

Automatic and accurate segmentation of colonic polyps can effectively help physicians quickly identify the size and location of polyps during endoscopy. However, due to the variable intestinal environment, polyps are highly similar to the surrounding tissue leading to over-segmentation or under-segmentation. To tackle these concerns, we propose a novel network with shape-awareness. Specifically, the network employs a dual-stream encoder, including the mainstream Transoformer-based encoder and an auxiliary encoder to extract global and local representations. The auxiliary encoder is built on deformable convolutional v3 for sensing the shape of polyps. In order to better identify the edges of polyps, the Edge Detection Guided Module (EDGM) is proposed to utilize edge detection operators to highlight the importance of edge information in low-level features. Furthermore, we introduce a Semantic Interaction Module (SIM) to better localize and calibrate targets by integrating global and local semantics in high-level features. Compared to other state-of-the-art methods, our model achieves better performance both in learning and generalization ability on five polyp segmentation datasets. The potential applicability of our model to other biomedical fields is demonstrated through its outstanding performance. The code is available at https://github.com/xff12138/EDO-SANet