Xingguang Zhou

Bio: Xingguang Zhou is an academic researcher from ETH Zurich. The author has contributed to research in topics: Mobile device & Image resolution. The author has an hindex of 4, co-authored 4 publications receiving 171 citations.

PIRM challenge on perceptual image enhancement on smartphones: Report

Abstract: This paper reviews the first challenge on efficient perceptual image enhancement with the focus on deploying deep learning models on smartphones. The challenge consisted of two tracks. In the first one, participants were solving the classical image super-resolution problem with a bicubic downscaling factor of 4. The second track was aimed at real-world photo enhancement, and the goal was to map low-quality photos from the iPhone 3GS device to the same photos captured with a DSLR camera. The target metric used in this challenge combined the runtime, PSNR scores and solutions’ perceptual results measured in the user study. To ensure the efficiency of the submitted models, we additionally measured their runtime and memory requirements on Android smartphones. The proposed solutions significantly improved baseline results defining the state-of-the-art for image enhancement on smartphones.

PIRM Challenge on Perceptual Image Enhancement on Smartphones: Report

Abstract: This paper reviews the first challenge on efficient perceptual image enhancement with the focus on deploying deep learning models on smartphones. The challenge consisted of two tracks. In the first one, participants were solving the classical image super-resolution problem with a bicubic downscaling factor of 4. The second track was aimed at real-world photo enhancement, and the goal was to map low-quality photos from the iPhone 3GS device to the same photos captured with a DSLR camera. The target metric used in this challenge combined the runtime, PSNR scores and solutions' perceptual results measured in the user study. To ensure the efficiency of the submitted models, we additionally measured their runtime and memory requirements on Android smartphones. The proposed solutions significantly improved baseline results defining the state-of-the-art for image enhancement on smartphones.

NTIRE 2019 Challenge on Image Enhancement: Methods and Results

Abstract: This paper reviews the first NTIRE challenge on perceptual image enhancement with the focus on proposed solutions and results. The participating teams were solving a real-world photo enhancement problem, where the goal was to map low-quality photos from the iPhone 3GS device to the same photos captured with Canon 70D DSLR camera. The considered problem embraced a number of computer vision subtasks, such as image denoising, image resolution and sharpness enhancement, image color/contrast/exposure adjustment, etc. The target metric used in this challenge combined PSNR and SSIM scores with solutions' perceptual results measured in the user study. The proposed solutions significantly improved baseline results, defining the state-of-the-art for practical image enhancement.

Range Scaling Global U-Net for Perceptual Image Enhancement on Mobile Devices

Abstract: Perceptual image enhancement on mobile devices—smart phones in particular—has drawn increasing industrial efforts and academic interests recently. Compared to digital single-lens reflex (DSLR) cameras, cameras on smart phones typically capture lower-quality images due to various hardware constraints. Without additional information, it is a challenging task to enhance the perceptual quality of a single image especially when the computation has to be done on mobile devices. In this paper we present a novel deep learning based approach—the Range Scaling Global U-Net (RSGUNet)—for perceptual image enhancement on mobile devices. Besides the U-Net structure that exploits image features at different resolutions, proposed RSGUNet learns a global feature vector as well as a novel range scaling layer that alleviate artifacts in the enhanced images. Extensive experiments show that the RSGUNet not only outputs enhanced images with higher subjective and objective quality, but also takes less inference time. Our proposal wins the 1st place by a great margin in track B of the Perceptual Image Enhancement on Smartphones Challenge (PRIM2018). Code is available at https://github.com/MTlab/ECCV-PIRM2018.

Research on the Planning of Civil Aviation Passenger Service Data Resource System

Abstract: The paper studies the current situation of service data distribution for civil aviation passenger, and expounds the planning methodology of passenger service data resource system. Firstly, it sorts out the process of air travel. Secondly, it maps the travel process with passenger behavior, access system and passenger service data. Based on the above analysis, the paper establishes the passenger service data resource system for civil aviation multi-agent and the whole travel process. Finally, it studies the interactive technology of passenger service data, and summarizes five representative interactive methods. This paper constructs the data resource system of civil aviation passenger service, which could effectively support the unification of data concept and the standardization of data service. Furthermore, it could provide data basis and decision-making reference for the improvement of passenger service quality.

Deep High-Resolution Representation Learning for Visual Recognition

Abstract: 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 \emph{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 \emph{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~{\url{this https URL}}.

Deep High-Resolution Representation Learning for Visual Recognition

Abstract: 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 .

Deep Learning for Image Super-Resolution: A Survey

Abstract: Image Super-Resolution (SR) is an important class of image processing techniqueso enhance the resolution of images and videos in computer vision. Recent years have witnessed remarkable progress of image super-resolution using deep learning techniques. This article aims to provide a comprehensive survey on recent advances of image super-resolution using deep learning approaches. In general, we can roughly group the existing studies of SR techniques into three major categories: supervised SR, unsupervised SR, and domain-specific SR. In addition, we also cover some other important issues, such as publicly available benchmark datasets and performance evaluation metrics. Finally, we conclude this survey by highlighting several future directions and open issues which should be further addressed by the community in the future.

AI Benchmark: Running Deep Neural Networks on Android Smartphones

Abstract: Over the last years, the computational power of mobile devices such as smartphones and tablets has grown dramatically, reaching the level of desktop computers available not long ago. While standard smartphone apps are no longer a problem for them, there is still a group of tasks that can easily challenge even high-end devices, namely running artificial intelligence algorithms. In this paper, we present a study of the current state of deep learning in the Android ecosystem and describe available frameworks, programming models and the limitations of running AI on smartphones. We give an overview of the hardware acceleration resources available on four main mobile chipset platforms: Qualcomm, HiSilicon, MediaTek and Samsung. Additionally, we present the real-world performance results of different mobile SoCs collected with AI Benchmark (http://ai-benchmark.com) that are covering all main existing hardware configurations.