Xu Jizheng

Bio: Xu Jizheng is an academic researcher from Microsoft. The author has contributed to research in topics: Video processing & Motion vector. The author has an hindex of 33, co-authored 457 publications receiving 5071 citations.

AOD-Net: All-in-One Dehazing Network

Abstract: This paper proposes an image dehazing model built with a convolutional neural network (CNN), called All-in-One Dehazing Network (AOD-Net). It is designed based on a re-formulated atmospheric scattering model. Instead of estimating the transmission matrix and the atmospheric light separately as most previous models did, AOD-Net directly generates the clean image through a light-weight CNN. Such a novel end-to-end design makes it easy to embed AOD-Net into other deep models, e.g., Faster R-CNN, for improving high-level tasks on hazy images. Experimental results on both synthesized and natural hazy image datasets demonstrate our superior performance than the state-of-the-art in terms of PSNR, SSIM and the subjective visual quality. Furthermore, when concatenating AOD-Net with Faster R-CNN, we witness a large improvement of the object detection performance on hazy images.

Efficient Parallel Framework for HEVC Motion Estimation on Many-Core Processors

Abstract: High Efficiency Video Coding (HEVC) provides superior coding efficiency than previous video coding standards at the cost of increasing encoding complexity. The complexity increase of motion estimation (ME) procedure is rather significant, especially when considering the complicated partitioning structure of HEVC. To fully exploit the coding efficiency brought by HEVC requires a huge amount of computations. In this paper, we analyze the ME structure in HEVC and propose a parallel framework to decouple ME for different partitions on many-core processors. Based on local parallel method (LPM), we first use the directed acyclic graph (DAG)-based order to parallelize coding tree units (CTUs) and adopt improved LPM (ILPM) within each CTU (DAGILPM), which exploits the CTU-level and prediction unit (PU)-level parallelism. Then, we find that there exist completely independent PUs (CIPUs) and partially independent PUs (PIPUs). When the degree of parallelism (DP) is smaller than the maximum DP of DAGILPM, we process the CIPUs and PIPUs, which further increases the DP. The data dependencies and coding efficiency stay the same as LPM. Experiments show that on a 64-core system, compared with serial execution, our proposed scheme achieves more than 30 and 40 times speedup for 1920 × 1080 and 2560 × 1600 video sequences, respectively.

A Highly Parallel Framework for HEVC Coding Unit Partitioning Tree Decision on Many-core Processors

Abstract: High Efficiency Video Coding (HEVC) uses a very flexible tree structure to organize coding units, which leads to a superior coding efficiency compared with previous video coding standards. However, such a flexible coding unit tree structure also places a great challenge for encoders. In order to fully exploit the coding efficiency brought by this structure, huge amount of computational complexity is needed for an encoder to decide the optimal coding unit tree for each image block. One way to achieve this is to use parallel computing enabled by many-core processors. In this paper, we analyze the challenge to use many-core processors to make coding unit tree decision. Through in-depth understanding of the dependency among different coding units, we propose a parallel framework to decide coding unit trees. Experimental results show that, on the Tile64 platform, our proposed method achieves averagely more than 11 and 16 times speedup for 1920x1080 and 2560x1600 video sequences, respectively, without any coding efficiency degradation.

Overview of the Emerging HEVC Screen Content Coding Extension

Abstract: A screen content coding (SCC) extension to High Efficiency Video Coding (HEVC) is currently under development by the Joint Collaborative Team on Video Coding, which is a joint effort from the ITU-T Video Coding Experts Group and the ISO/IEC Moving Picture Experts Group. The main goal of the HEVC-SCC standardization effort is to enable significantly improved compression performance for videos containing a substantial amount of still or moving rendered graphics, text, and animation rather than, or in addition to, camera-captured content. This paper provides an overview of the technical features and characteristics of the current HEVC-SCC test model and related coding tools, including intra-block copy, palette mode, adaptive color transform, and adaptive motion vector resolution. The performance of the SCC extension is compared against existing standards in terms of bitrate savings at equal distortion.

Comparison of the Coding Efficiency of Video Coding Standards—Including High Efficiency Video Coding (HEVC)

Abstract: The compression capability of several generations of video coding standards is compared by means of peak signal-to-noise ratio (PSNR) and subjective testing results. A unified approach is applied to the analysis of designs, including H.262/MPEG-2 Video, H.263, MPEG-4 Visual, H.264/MPEG-4 Advanced Video Coding (AVC), and High Efficiency Video Coding (HEVC). The results of subjective tests for WVGA and HD sequences indicate that HEVC encoders can achieve equivalent subjective reproduction quality as encoders that conform to H.264/MPEG-4 AVC when using approximately 50% less bit rate on average. The HEVC design is shown to be especially effective for low bit rates, high-resolution video content, and low-delay communication applications. The measured subjective improvement somewhat exceeds the improvement measured by the PSNR metric.

DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks

Abstract: We present DeblurGAN, an end-to-end learned method for motion deblurring. The learning is based on a conditional GAN and the content loss. DeblurGAN achieves state-of-the art performance both in the structural similarity measure and visual appearance. The quality of the deblurring model is also evaluated in a novel way on a real-world problem - object detection on (de-)blurred images. The method is 5 times faster than the closest competitor - Deep-Deblur [25]. We also introduce a novel method for generating synthetic motion blurred images from sharp ones, allowing realistic dataset augmentation. The model, code and the dataset are available at https://github.com/KupynOrest/DeblurGAN

Benchmarking Single-Image Dehazing and Beyond

Abstract: We present a comprehensive study and evaluation of existing single-image dehazing algorithms, using a new large-scale benchmark consisting of both synthetic and real-world hazy images, called REalistic Single-Image DEhazing (RESIDE). RESIDE highlights diverse data sources and image contents, and is divided into five subsets, each serving different training or evaluation purposes. We further provide a rich variety of criteria for dehazing algorithm evaluation, ranging from full-reference metrics to no-reference metrics and to subjective evaluation, and the novel task-driven evaluation. Experiments on RESIDE shed light on the comparisons and limitations of the state-of-the-art dehazing algorithms, and suggest promising future directions.