Video-SwinUNet: Spatio-temporal Deep Learning Framework for VFSS Instance Segmentation

• Published in: 2023 IEEE International Conference on Image Processing (ICIP) pp. 2470 - 2474
• Main Authors: Zeng, Chengxi, Yang, Xinyu, Smithard, David, Mirmehdi, Majid, Gambaruto, Alberto M, Burghardt, Tilo
• Format: Conference Proceeding
• Language: English
• Published: IEEE 08.10.2023
• Subjects: Deep learning; Dysphagia; Feature extraction; Image segmentation; Neural networks; Performance gain; Pipelines; Semantics; Swin Transformer; Video Tracking; Videofluoroscopy

This paper presents a deep learning framework for medical video segmentation. Convolution neural network (CNN) and transformer-based methods have achieved great milestones in medical image segmentation tasks due to their incredible semantic feature encoding and global information comprehension abili...