Blind Video Quality Assessment Research Articles

SR4KVQA: Video quality assessment database and metric for 4K super-resolution

The quality assessment for 4K super-resolution (SR) videos can be conducive to the optimization of video SR algorithms. To improve the subjective and objective consistency of the SR quality assessment, a 4K video database and a blind metric are proposed in this paper. In the database SR4KVQA, there are 30 4K pristine videos, from which 600 SR 4K distorted videos with mean opinion score (MOS) labels are generated by three classic interpolation methods, six SR algorithms based on the deep neural network (DNN), and two SR algorithms based on the generative adversarial network (GAN). The benchmark experiment of the proposed database indicates that video quality assessment (VQA) of the 4K SR videos is challenging for the existing metrics. Among those metrics, the Video-Swin-Transformer backbone demonstrates tremendous potential in the VQA task. Accordingly, a blind VQA metric based on the Video-Swin-Transformer backbone is established, where the normalized loss function and optimized spatio-temporal sampling strategy are applied. The experiment result manifests that the Pearson linear correlation coefficient (PLCC) and Spearman rank-order correlation coefficient (SROCC) of the proposed metric reach 0.8011 and 0.8275 respectively on the SR4KVQA database, which outperforms or competes with the state-of-the-art VQA metrics. The database and the code proposed in this paper are available in the GitHub repository, https://github.com/AlexReadyNico/SR4KVQA.

CoSTA: Co-training spatial–temporal attention for blind video quality assessment

Self-attention-based Transformer has achieved great success in many computer vision tasks. However, its application to blind video quality assessment (VQA) is far from comprehensive. Evaluating the quality of in-the-wild videos is challenging due to the unknown of pristine reference and shooting distortion. This paper presents a Co-trained Space–Time Attention network for the blind VQA problem, termed CoSTA. Specifically, we first build CoSTA by alternately concatenating the divided space–time attention. Then, to facilitate the training of CoSTA, we design a vectorized regression loss by encoding the mean opinion score (MOS) to the probability vector and embedding a special token as the learnable variable of MOS, leading to the better fitting of the human rating process. Finally, to solve the data-hungry problem within Transformer, we propose to co-train the spatial and temporal attention weights using both images and videos. Various experiments are conducted on the de-facto in-the-wild video datasets, including LIVE-Qualcomm, LIVE-VQC, KoNViD-1k, YouTube-UGC, LSVQ, LSVQ-1080p, and DVL2021. Experimental results demonstrate the superiority of the proposed CoSTA over the state-of-the-art. The source code is publicly available at https://github.com/GZHU-DVL/CoSTA.

Multi-Modal Learning-Based Blind Video Quality Assessment Metric for Synthesized Views

Multi-Modal Learning-Based Blind Video Quality Assessment Metric for Synthesized Views

Blind video quality assessment based on Spatio-Temporal Feature Resolver

As a method to measure video quality, blind video quality assessment (BVQA) plays an important role in video related applications. Feature extraction, as a vital component of BVQA, significantly impacts the performance and speed of the method. In the process of designing BVQA method, we make the feature extraction process independent of the whole quality evaluation method. First, a Spatio-Temporal Feature Resolver (STFR) is obtained by training. Then, the STFR is employed to directly extract the spatio-temporal features of the video sequences. Finally, the extracted features are mapped to quality scores using Support Vector Regression (SVR). STFR only needs to be trained in an undistorted video sequence, and can be directly applied to video sequences of various scenes, which has a universality. To evaluate the effectiveness of the proposed method, the experimental results on four published video quality databases show that the proposed BVQA method not only achieves more accurately than the existing BVQA methods, but also exhibits significant competitiveness in computational speed.

FAVER: Blind quality prediction of variable frame rate videos

Video quality assessment (VQA) remains an important and challenging problem that affects many applications at the widest scales. Recent advances in mobile devices and cloud computing techniques have made it possible to capture, process, and share high resolution, high frame rate (HFR) videos across the Internet nearly instantaneously. Being able to monitor and control the quality of these streamed videos can enable the delivery of more enjoyable content and perceptually optimized rate control. Accordingly, there is a pressing need to develop VQA models that can be deployed at enormous scales. While some recent effects have been applied to full-reference (FR) analysis of variable frame rate and HFR video quality, the development of no-reference (NR) VQA algorithms targeting frame rate variations has been little studied. Here, we propose a first-of-a-kind blind VQA model for evaluating HFR videos, which we dub the Framerate-Aware Video Evaluator w/o Reference (FAVER). FAVER uses extended models of spatial natural scene statistics that encompass space–time wavelet-decomposed video signals, and leverages the advantages of the deep neural network to provide motion perception, to conduct efficient frame rate sensitive quality prediction. Our extensive experiments on several HFR video quality datasets show that FAVER outperforms other blind VQA algorithms at a reasonable computational cost. To facilitate reproducible research and public evaluation, an implementation of FAVER is being made freely available online: https://github.com/uniqzheng/HFR-BVQA.

2BiVQA: Double Bi-LSTM-based Video Quality Assessment of UGC Videos

Recently, with the growing popularity of mobile devices as well as video sharing platforms (e.g., YouTube, Facebook, TikTok, and Twitch), User-Generated Content (UGC) videos have become increasingly common and now account for a large portion of multimedia traffic on the internet. Unlike professionally generated videos produced by filmmakers and videographers, typically, UGC videos contain multiple authentic distortions, generally introduced during capture and processing by naive users. Quality prediction of UGC videos is of paramount importance to optimize and monitor their processing in hosting platforms, such as their coding, transcoding, and streaming. However, blind quality prediction of UGC is quite challenging, because the degradations of UGC videos are unknown and very diverse, in addition to the unavailability of pristine reference. Therefore, in this article, we propose an accurate and efficient Blind Video Quality Assessment (BVQA) model for UGC videos, which we name 2BiVQA for double Bi-LSTM Video Quality Assessment. 2BiVQA metric consists of three main blocks, including a pre-trained Convolutional Neural Network to extract discriminative features from image patches, which are then fed into two Recurrent Neural Networks for spatial and temporal pooling. Specifically, we use two Bi-directional Long Short-term Memory networks, the first is used to capture short-range dependencies between image patches, while the second allows capturing long-range dependencies between frames to account for the temporal memory effect. Experimental results on recent large-scale UGC VQA datasets show that 2BiVQA achieves high performance at lower computational cost than most state-of-the-art VQA models. The source code of our 2BiVQA metric is made publicly available at https://github.com/atelili/2BiVQA .

Blind Video Quality Assessment for Ultra-High-Definition Video Based on Super-Resolution and Deep Reinforcement Learning.

Ultra-high-definition (UHD) video has brought new challenges to objective video quality assessment (VQA) due to its high resolution and high frame rate. Most existing VQA methods are designed for non-UHD videos-when they are employed to deal with UHD videos, the processing speed will be slow and the global spatial features cannot be fully extracted. In addition, these VQA methods usually segment the video into multiple segments, predict the quality score of each segment, and then average the quality score of each segment to obtain the quality score of the whole video. This breaks the temporal correlation of the video sequences and is inconsistent with the characteristics of human visual perception. In this paper, we present a no-reference VQA method, aiming to effectively and efficiently predict quality scores for UHD videos. First, we construct a spatial distortion feature network based on a super-resolution model (SR-SDFNet), which can quickly extract the global spatial distortion features of UHD videos. Then, to aggregate the spatial distortion features of each UHD frame, we propose a time fusion network based on a reinforcement learning model (RL-TFNet), in which the actor network continuously combines multiple frame features extracted by SR-SDFNet and outputs an action to adjust the current quality score to approximate the subjective score, and the critic network outputs action values to optimize the quality perception of the actor network. Finally, we conduct large-scale experiments on UHD VQA databases and the results reveal that, compared to other state-of-the-art VQA methods, our method achieves competitive quality prediction performance with a shorter runtime and fewer model parameters.

End-to-End Blind Video Quality Assessment Based on Visual and Memory Attention Modeling

End-to-End Blind Video Quality Assessment Based on Visual and Memory Attention Modeling

Blindly Assess Quality of In-the-Wild Videos via Quality-Aware Pre-Training and Motion Perception

Perceptual quality assessment of the videos acquired in the wilds is of vital importance for quality assurance of video services. The inaccessibility of reference videos with pristine quality and the complexity of authentic distortions pose great challenges for this kind of blind video quality assessment (BVQA) task. Although model-based transfer learning is an effective and efficient paradigm for the BVQA task, it remains to be a challenge to explore <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">what</i> and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">how</i> to bridge the domain shifts for better video representation. In this work, we propose to transfer knowledge from image quality assessment (IQA) databases with authentic distortions and large-scale action recognition with rich motion patterns. We rely on both groups of data to learn the feature extractor and use a mixed list-wise ranking loss function to train the entire model on the target VQA databases. Extensive experiments on six benchmarking databases demonstrate that our method performs very competitively under both individual database and mixed databases training settings. We also verify the rationality of each component of the proposed method and explore a simple ensemble trick for further improvement.

Spatiotemporal Representation Learning for Blind Video Quality Assessment

Blind video quality assessment (BVQA) is of great importance for video-related applications, yet still challenging even in this deep learning era. The difficulty lies in the shortage of large-scale labeled data, thus making it hard to train a robust spatiotemporal encoder for BVQA. To relieve such difficulty, we first build a video dataset, which contains over 320K samples suffering from various compression and transmission artifacts. While manually annotating the dataset with subjective perception is much labor-intensive and time-consuming, we adopt reference-based VQA algorithms to weakly label the data automatically. We consider that single weak label is derived from single knowledge, which is deficient and incomplete for VQA. To alleviate the bias from single weak label (i.e., single knowledge) in the weakly labeled dataset, we propose HEterogeneous Knowledge Ensemble (HEKE) for spatiotemporal representation learning. Compared to learning from single knowledge, learning with HEKE is thought to achieve a lower infimum theoretically, and obtain richer representation. On the basis of the built dataset and the HEKE methodology, a feature encoder specific to BVQA is formed, and directly extract spatiotemporal representation from videos. Then, the video quality can be either acquired in a completely BVQA manner without ground truth, or via a finetuning-based regressor with labels. Extensive experiments on various VQA databases show that our BVQA model with the pretrained encoder achieves the state-of-the-art performance. More surprisingly, even trained on the synthetic data, our model still shows competitive performance on authentic databases. The data and source code will be available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/Sissuire/BVQA-HEKE</uri> .

A continuous bitstream-based blind video quality assessment using multi-layer perceptron

We propose a continuous blind/no-reference video quality assessment (NR-VQA) algorithm based on features extracted from the bitstream, i.e., without decoding the video. The resulting algorithm requires minimal training and adopts a simple multi-layer perceptron for score prediction. The algorithm is computationally appealing. To assess the performance of the algorithm, both the Pearson Correlation Coefficient (PCC) and the Spearman Rank Ordered Correlation Coefficient (SROCC) are computed between the predicted values and the quality scores of two databases. The proposed approach is shown to have a high correlation with human visual perception of quality.

Blind video quality assessment based on multilevel video perception

Blind video quality assessment (VQA) metrics predict the quality of videos without the presence of reference videos. This paper proposes a new blind VQA model based on multilevel video perception, abbreviated as MVP. The model fuses three levels of video features occurring in natural video scenes to predict video quality: natural video statistics (NVS) features, global motion features and motion temporal correlation features. They represent video scene characteristics, video motion types, and video temporal correlation variations. In the process of motion feature extraction, motion compensation filtering video enhancement is adopted to highlight the motion characteristics of videos so as to improve the perceptual correlations of the video features. The experimental results on the LIVE and CSIQ video databases show that the predicted video scores of the new model are highly correlated with human perception and have low root mean square errors. MVP obviously outperforms state-of-art blind VQA metrics, and particularly demonstrates competitive performance even compared against top-performing full reference VQA metrics.

A New Blind Video Quality Metric for Assessing Different Turbulence Mitigation Algorithms

Although many algorithms have been proposed to mitigate air turbulence in optical videos, there do not seem to be consistent blind video quality assessment metrics that can reliably assess different approaches. Blind video quality assessment metrics are necessary because many videos containing air turbulence do not have ground truth. In this paper, a simple and intuitive blind video quality assessment metric is proposed. This metric can reliably and consistently assess various turbulent mitigation algorithms for optical videos. Experimental results using more than 10 videos in the literature show that the proposed metrics correlate well with human subjective evaluations. Compared with an existing blind video metric and two other blind image quality metrics, the proposed metrics performed consistently better.

No-reference screen content video quality assessment

How to effectively and accurately measure the degradation of media content is an important research topic in the field of image or video processing. Application scenarios such as online meetings, distance learning, and live game streaming make screen content video become a hot spot in Video Quality Assessment (VQA) research. However, to the best of our knowledge, there is currently no no-reference VQA model designed specifically for screen content videos. In this paper, we propose a blind VQA model for screen content videos. This model first uses a multi-scale approach to extract several groups of features, including gradient features, relative standard deviation features, compression features, frequency domain features and inter-frame features. Through training with labeled videos, the model then uses support vector regressor to map the frame feature vectors to video quality scores. We validate the model on the CSCVQ database. Experiments show that our proposed model outperforms the existing full- and no-reference quality evaluation metrics and is also competitive in terms of stability and computational efficiency.

HDR-BVQM: High dynamic range blind video quality model

Considerable progress has been made toward developing standard dynamic range (SDR) blind video quality assessment (BVQA) models that do not require any baseline reference for quality prediction. However, there is no such method for the high dynamic range (HDR) content. Unlike SDR video, HDR video represents a high-fidelity representation of the real-world scene by preserving the wide luminance range and color gamut. Therefore, SDR BVQA models are not suitable for HDR BVQA. Towards ameliorating this, a first-of-its-kind BVQA model for HDR content is presented in this work. The proposed HDR blind video quality model (HDR-BVQM) is inspired by the spatio-temporal natural scene statistics model, previously employed in SDR blind quality assessment metrics. To build our proposed model, we first develop a comprehensive subjective HDR video quality dataset, including 228 distorted videos generated through three different (H.264, HEVC, packet drop) distortion processes from 19 pristine HDR videos. The developed dataset is then used to extract HDR relevant features, which vary in different distortion types, to train and test the proposed HDR-BVQM. The features are based on the pointwise, pairwise log-derivative, and motion coherence based statistics. Finally, detailed validation and performance comparison is performed with full-reference HDR and no-reference SDR quality assessment methods. The results reveal that the quality prediction by HDR-BVQM correlates with the human judgment of quality.

Blind Video Quality Assessment With Weakly Supervised Learning and Resampling Strategy

Due to the 3D spatiotemporal regularities of natural videos and small-scale video quality databases, effective objective video quality assessment (VQA) metrics are difficult to obtain but highly desirable. In this paper, we propose a general-purpose no-reference VQA framework that is based on weakly supervised learning with a convolutional neural network (CNN) and a resampling strategy. First, an eight-layer CNN is trained by weakly supervised learning to construct the relationship between the deformations of the 3D discrete cosine transform of video blocks and the corresponding weak labels judged by a full-reference (FR) VQA metric. Thus, the CNN obtains the quality assessment capacity converted from the FR-VQA metric, and the effective features of the distorted videos can be extracted through the trained network. Then, we map the frequency histogram calculated from the quality score vectors predicted by the trained network onto the perceptual quality. Especially, to improve the performance of the mapping function, we transfer the frequency histogram of the distorted images and videos to resample the training set. The experiments are carried out on several widely used VQA databases. The experimental results demonstrate that the proposed method is on a par with some state-of-the-art VQA metrics and has promising robustness.

A Completely Blind Video Integrity Oracle.

Considerable progress has been made toward developing still picture perceptual quality analyzers that do not require any reference picture and that are not trained on human opinion scores of distorted images. However, there do not yet exist any such completely blind video quality assessment (VQA) models. Here, we attempt to bridge this gap by developing a new VQA model called the video intrinsic integrity and distortion evaluation oracle (VIIDEO). The new model does not require the use of any additional information other than the video being quality evaluated. VIIDEO embodies models of intrinsic statistical regularities that are observed in natural vidoes, which are used to quantify disturbances introduced due to distortions. An algorithm derived from the VIIDEO model is thereby able to predict the quality of distorted videos without any external knowledge about the pristine source, anticipated distortions, or human judgments of video quality. Even with such a paucity of information, we are able to show that the VIIDEO algorithm performs much better than the legacy full reference quality measure MSE on the LIVE VQA database and delivers performance comparable with a leading human judgment trained blind VQA model. We believe that the VIIDEO algorithm is a significant step toward making real-time monitoring of completely blind video quality possible.

Blind prediction of natural video quality.

We propose a blind (no reference or NR) video quality evaluation model that is nondistortion specific. The approach relies on a spatio-temporal model of video scenes in the discrete cosine transform domain, and on a model that characterizes the type of motion occurring in the scenes, to predict video quality. We use the models to define video statistics and perceptual features that are the basis of a video quality assessment (VQA) algorithm that does not require the presence of a pristine video to compare against in order to predict a perceptual quality score. The contributions of this paper are threefold. 1) We propose a spatio-temporal natural scene statistics (NSS) model for videos. 2) We propose a motion model that quantifies motion coherency in video scenes. 3) We show that the proposed NSS and motion coherency models are appropriate for quality assessment of videos, and we utilize them to design a blind VQA algorithm that correlates highly with human judgments of quality. The proposed algorithm, called video BLIINDS, is tested on the LIVE VQA database and on the EPFL-PoliMi video database and shown to perform close to the level of top performing reduced and full reference VQA algorithms.