Guided by the free-energy principle, generative adversarial networks (GAN)-based no-reference image quality assessment (NR-IQA) methods have improved the image quality prediction accuracy. However, the GAN cannot well handle the restoration task for the free-energy principle-guided NR-IQA methods, especially for the severely destroyed images, which results in that the quality reconstruction relationship between the distorted image and its restored image cannot be accurately built. To address this problem, a visual compensation restoration network (VCRNet)-based NR-IQA method is proposed, which uses a non-adversarial model to efficiently handle the distorted image restoration task. The proposed VCRNet consists of a visual restoration network and a quality estimation network. To accurately build the quality reconstruction relationship between the distorted image and its restored image, a visual compensation module, an optimized asymmetric residual block, and an error map-based mixed loss function, are proposed for increasing the restoration capability of the visual restoration network. For further addressing the NR-IQA problem of severely destroyed images, the multi-level restoration features which are obtained from the visual restoration network are used for the image quality estimation. To prove the effectiveness of the proposed VCRNet, seven representative IQA databases are used, and experimental results show that the proposed VCRNet achieves the state-of-the-art image quality prediction accuracy. The implementation of the proposed VCRNet has been released at https://github.com/NUIST-Videocoding/VCRNet.

Recently, the Generative Adversarial Networks (GANs) are fast becoming a key promising research direction in computational intelligence. To improve the modeling ability of GANs, loss functions are used to measure the differences between samples generated by the model and real samples, and make the model learn towards the goal. In this paper, we perform a survey for the loss functions used in GANs, and analyze the pros and cons of these loss functions. Firstly, the basic theory of GANs, and its training mechanism are introduced. Then, the loss functions used in GANs are summarized, including not only the objective functions of GANs, but also the application-oriented GANs' loss functions. Thirdly, the experiments and analyses of representative loss functions are discussed. Finally, several suggestions on how to choose appropriate loss functions in a specific task are given.

Deep neural networks have achieved great performance on blind Image Quality Assessment (IQA), but it is still challenging for using one network to accurately predict the quality of images with different distortions. In this paper, a Distortion-Aware Convolutional Neural Network (DACNN) is proposed for blind IQA, which works effectively for not only synthetically distorted images but also authentically distorted images. The proposed DACNN consists of a distortion aware module, a distortion fusion module, and a quality prediction module. In the distortion aware module, a Siamese network-based pretraining strategy is proposed to design a synthetic distortion-aware network for full learning the synthetic distortions, and an authentic distortion-aware network is used for extracting the authentic distortions. To efficiently fuse the learned distortion features, and make the network pay more attention to the essential features, a weight-adaptive fusion network is proposed to adaptively adjust the weight of each distortion. Finally, the quality prediction module is adopted to map the fused features to a quality score. Extensive experiments on four authentic IQA databases and four synthetic IQA databases have proved the effectiveness of the proposed DACNN.

The raw video data can be compressed much by the latest video coding standard, high efficiency video coding (HEVC). However, the block-based hybrid coding used in HEVC will incur lots of artifacts in compressed videos, the video quality will be severely influenced. To settle this problem, the in-loop filtering is used in HEVC to eliminate artifacts. Inspired by the success of deep learning, we propose an efficient in-loop filtering algorithm based on the enhanced deep convolutional neural networks (EDCNN) for significantly improving the performance of in-loop filtering in HEVC. Firstly, the problems of traditional convolutional neural networks models, including the normalization method, network learning ability, and loss function, are analyzed. Then, based on the statistical analyses, the EDCNN is proposed for efficiently eliminating the artifacts, which adopts three solutions, including a weighted normalization method, a feature information fusion block, and a precise loss function. Finally, the PSNR enhancement, PSNR smoothness, RD performance, subjective test, and computational complexity/GPU memory consumption are employed as the evaluation criteria, and experimental results show that when compared with the filter in HM16.9, the proposed in-loop filtering algorithm achieves an average of 6.45% BDBR reduction and 0.238 dB BDPSNR gains.

Screen content coding (SCC) has evolved into the extension of the High Efficiency Video Coding (HEVC). Low-latency, real-time transport between devices in the form of screen content video is becoming popular in many applications. However, the complexity of encoder is still very high for intra prediction in HEVC-based SCC. This paper proposes a fast intra prediction method based on content property analysis for HEVC-based SCC. First, coding units (CUs) are classified into natural content CUs (NCCUs) and screen content CUs (SCCUs), based on the statistic characteristics of the content. For NCCUs, the newly adopted prediction modes, including intra block copy mode and palette mode are skipped, if the DC or PLANR mode is the best mode, after testing the traditional intra prediction rough modes. In addition, the quadtree partition process is also terminated due to the homogeneous and smooth block usually chooses a large size CU. For SCCUs, a rank-based decision strategy is introduced to terminate the splitting process of current CU. For all CUs, the bit per pixel of current CU is used to make a CU size decision. Meanwhile, the depth information of neighboring CUs and co-located CU are utilized to further improve the performance. Experimental results show that the proposed algorithm can save 44.92% encoding time on average with negligible loss of video quality.