BiO-Net: Learning Recurrent Bi-directional Connections for Encoder-Decoder Architecture
This work addresses the problem of high computational cost in medical image segmentation and related tasks for researchers and practitioners, offering a parameter-efficient improvement.
The paper tackles the issue of increased model complexity in U-Net variants by proposing BiO-Net, which uses recurrent bi-directional skip connections to enhance encoder-decoder architectures without extra parameters, and it significantly outperforms vanilla U-Net and other state-of-the-art methods on various medical image analysis tasks.
U-Net has become one of the state-of-the-art deep learning-based approaches for modern computer vision tasks such as semantic segmentation, super resolution, image denoising, and inpainting. Previous extensions of U-Net have focused mainly on the modification of its existing building blocks or the development of new functional modules for performance gains. As a result, these variants usually lead to an unneglectable increase in model complexity. To tackle this issue in such U-Net variants, in this paper, we present a novel Bi-directional O-shape network (BiO-Net) that reuses the building blocks in a recurrent manner without introducing any extra parameters. Our proposed bi-directional skip connections can be directly adopted into any encoder-decoder architecture to further enhance its capabilities in various task domains. We evaluated our method on various medical image analysis tasks and the results show that our BiO-Net significantly outperforms the vanilla U-Net as well as other state-of-the-art methods. Our code is available at https://github.com/tiangexiang/BiO-Net.