Real-Time Image Segmentation via Hybrid Convolutional-Transformer Architecture Search

arXiv:2403.10413v1 Announce Type: new
Abstract: Image segmentation is one of the most fundamental problems in computer vision and has drawn a lot of attentions due to its vast applications in image understanding and autonomous driving. However, designing effective and efficient segmentation neural architectures is a labor-intensive process that may require lots of trials by human experts. In this paper, we address the challenge of integrating multi-head self-attention into high resolution representation CNNs efficiently, by leveraging architecture search. Manually replacing convolution …

architecture arxiv cs.cv hybrid image real-time search segmentation transformer transformer architecture type via