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; (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. ! 1 INTRODUCTION D EEP convolutional neural networks (DCNNs) have achieved state-of-the-art results in many computer vision tasks, such as image classification, object detection, semantic segmentation, human pose estimation, and so on. The strength is that DCNNs are able to learn richer representations than conventional hand-crafted representations. Most recently-developed classification networks, including AlexNet [59], VGGNet [101], GoogleNet [108], ResNet [39], etc., follow the design rule of LeNet-5 [61]. This is depicted in Figure 1 (a): gradually reduce the spatial size of the feature maps, connect the convolutions from high resolution to low resolution in series, and lead to a low-resolution representation, which is further processed for classification.High-resolution representations are needed for positionsensitive tasks, e.g., semantic segmentation, human pose estimation, and object detection. The previous state-of-the-art methods adopt the high-resolution recovery process to raise the representation resolution from the low-resolution representation outputted by a classification or classification-like network as depicted in Figure 1 (b), e.g., Hourglass [83], Seg-Net [3], DeconvNet [85], U-Net [95], SimpleBaseline [124], and encoder-decoder [90]. In addition, dilated convolutions are used to remove some down-sample layers and thus yield medium-resolution representations [15], [144].We present a novel architecture, namely High-Resolution Net (HRNet), which is able to maintain high-resolution representations through the whole process. We start from a highresolution convolution stream, gradually add high-to-low resolution convolution streams one by one, and connect the multi-resolution streams in parallel. The resulting network • J. Wang is with Microsoft Research,
