Deep High-Resolution Representation Learning for Visual Recognition

• Published in: IEEE transactions on pattern analysis and machine intelligence Vol. 43; no. 10; pp. 3349 - 3364
• Main Authors: Wang, Jingdong, Sun, Ke, Cheng, Tianheng, Jiang, Borui, Deng, Chaorui, Zhao, Yang, Liu, Dong, Mu, Yadong, Tan, Mingkui, Wang, Xinggang, Liu, Wenyu, Xiao, Bin
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.10.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Computer vision; Convolutional codes; High resolution; high-resolution representations; HRNet; human pose estimation; Image resolution; Image segmentation; Indexes; low-resolution representations; Object detection; Object recognition; Pose estimation; Position sensing; Representations; Semantic segmentation; Semantics; Spatial resolution
• ISSN: 0162-8828; 1939-3539; 2160-9292; 1939-3539
• DOI: 10.1109/TPAMI.2020.2983686

High-resolution representations are essential for position-sensitive vision problems, such as human pose estimation, semantic segmentation, and object detection. Existing state-of-the-art frameworks first encode the input image as a low-resolution representation through a subnetwork that is formed by connecting high-to-low resolution convolutions in series (e.g., ResNet, VGGNet), and then recover the high-resolution representation from the encoded low-resolution representation. Instead, our proposed network, named as High-Resolution Network (HRNet), maintains high-resolution representations through the whole process. There are two key characteristics: (i) Connect the high-to-low resolution convolution streams in parallel and (ii) repeatedly exchange the information across resolutions. The benefit is that the resulting representation is semantically richer and spatially more precise. We show the superiority of the proposed HRNet in a wide range of applications, including human pose estimation, semantic segmentation, and object detection, suggesting that the HRNet is a stronger backbone for computer vision problems. All the codes are available at https://github.com/HRNet .