Dmitriy Vatolin

Ivan Molodetskikh, Kirill Malyshev, Mark Mirgaleev et al.

Modern image super-resolution methods generate detailed, visually appealing results, but they often introduce visual artifacts: unnatural patterns and texture distortions that degrade perceived quality. These defects vary widely in perceptual impact--some are barely noticeable, while others are highly disturbing--yet existing detection methods treat them equally. We propose artifact prominence as an evaluative target, defined as the fraction of viewers who judge a highlighted region to contain a noticeable artifact. We design a crowdsourced annotation protocol and construct SR-Prominence, a dataset suite containing 3,935 artifact masks from DeSRA, Open Images, Urban100, and a realistic no-ground-truth Urban100-HR setting, annotated with prominence. Re-annotating DeSRA reveals that 48.2% of its in-lab binary artifacts are not noticed by a majority of viewers. Across the suite, we audit SR artifact detectors, image-quality metrics, and SR methods. We find that classical full-reference metrics, especially SSIM and DISTS, provide surprisingly strong localized prominence signals, whereas no-reference IQA methods and specialized artifact detectors often fail to generalize across datasets and reference settings. SR-Prominence is released with an objective scoring protocol that allows new metrics to be benchmarked on our suite without further crowdsourcing. Together, the data and protocols enable SR artifact evaluation to move from binary defect presence toward perceptual impact. SR-Prominence is available at https://huggingface.co/datasets/imolodetskikh/sr-artifact-prominence.

Viacheslav Meshchaninov, Ivan Molodetskikh, Dmitriy Vatolin

Upscaled video detection is a helpful tool in multimedia forensics, but it is a challenging task that involves various upscaling and compression algorithms. There are many resolution-enhancement methods, including interpolation and deep-learning-based super-resolution, and they leave unique traces. In this work, we propose a new upscaled-resolution-detection method based on learning of visual representations using contrastive and cross-entropy losses. To explain how the method detects videos, we systematically review the major components of our framework - in particular, we show that most data-augmentation approaches hinder the learning of the method. Through extensive experiments on various datasets, we demonstrate that our method effectively detects upscaling even in compressed videos and outperforms the state-of-the-art alternatives. The code and models are publicly available at https://github.com/msu-video-group/SRDM

Anastasia Antsiferova, Sergey Lavrushkin, Maksim Smirnov et al.

Video-quality measurement is a critical task in video processing. Nowadays, many implementations of new encoding standards - such as AV1, VVC, and LCEVC - use deep-learning-based decoding algorithms with perceptual metrics that serve as optimization objectives. But investigations of the performance of modern video- and image-quality metrics commonly employ videos compressed using older standards, such as AVC. In this paper, we present a new benchmark for video-quality metrics that evaluates video compression. It is based on a new dataset consisting of about 2,500 streams encoded using different standards, including AVC, HEVC, AV1, VP9, and VVC. Subjective scores were collected using crowdsourced pairwise comparisons. The list of evaluated metrics includes recent ones based on machine learning and neural networks. The results demonstrate that new no-reference metrics exhibit a high correlation with subjective quality and approach the capability of top full-reference metrics.

Ekaterina Shumitskaya, Anastasia Antsiferova, Dmitriy Vatolin

Universal adversarial perturbation attacks are widely used to analyze image classifiers that employ convolutional neural networks. Nowadays, some attacks can deceive image- and video-quality metrics. So sustainability analysis of these metrics is important. Indeed, if an attack can confuse the metric, an attacker can easily increase quality scores. When developers of image- and video-algorithms can boost their scores through detached processing, algorithm comparisons are no longer fair. Inspired by the idea of universal adversarial perturbation for classifiers, we suggest a new method to attack differentiable no-reference quality metrics through universal perturbation. We applied this method to seven no-reference image- and video-quality metrics (PaQ-2-PiQ, Linearity, VSFA, MDTVSFA, KonCept512, Nima and SPAQ). For each one, we trained a universal perturbation that increases the respective scores. We also propose a method for assessing metric stability and identify the metrics that are the most vulnerable and the most resistant to our attack. The existence of successful universal perturbations appears to diminish the metric's ability to provide reliable scores. We therefore recommend our proposed method as an additional verification of metric reliability to complement traditional subjective tests and benchmarks.

Anastasia Antsiferova, Khaled Abud, Aleksandr Gushchin et al.

Nowadays, neural-network-based image- and video-quality metrics perform better than traditional methods. However, they also became more vulnerable to adversarial attacks that increase metrics' scores without improving visual quality. The existing benchmarks of quality metrics compare their performance in terms of correlation with subjective quality and calculation time. Nonetheless, the adversarial robustness of image-quality metrics is also an area worth researching. This paper analyses modern metrics' robustness to different adversarial attacks. We adapted adversarial attacks from computer vision tasks and compared attacks' efficiency against 15 no-reference image- and video-quality metrics. Some metrics showed high resistance to adversarial attacks, which makes their usage in benchmarks safer than vulnerable metrics. The benchmark accepts submissions of new metrics for researchers who want to make their metrics more robust to attacks or to find such metrics for their needs. The latest results can be found online: https://videoprocessing.ai/benchmarks/metrics-robustness.html.

Artem Borisov, Evgeney Bogatyrev, Khaled Abud et al.

Super-Resolution (SR) has advanced rapidly in recent years, with diffusion-based models achieving unprecedented fidelity at the cost of introducing new types of visual artifacts. While existing Image Quality Assessment (IQA) methods provide holistic quality scores, they lack interpretability and fail to distinguish between different artifact types arising from modern SR approaches. To address this gap, we introduce SR-Ground, a large-scale dataset specifically designed for fine-grained artifact segmentation in super-resolved images. The dataset comprises images processed by a diverse set of state-of-the-art SR models, with pixel-level annotations for multiple artifact categories. We conduct a large-scale crowdsourcing study involving 1,062 participants to validate and refine automatically generated segmentations, resulting in a high-quality dataset of 63,000 images spanning 6 distinct artifact types. We demonstrate that training IQA models with grounding capabilities on SR-Ground significantly improves performance on downstream tasks. Furthermore, we introduce a fine-tuning pipeline that leverages our grounding model to reduce perceptible artifacts in SR outputs, showcasing the practical utility of our dataset.

Alexander Gushchin, Khaled Abud, Georgii Bychkov et al.

In the field of Image Quality Assessment (IQA), the adversarial robustness of the metrics poses a critical concern. This paper presents a comprehensive benchmarking study of various defense mechanisms in response to the rise in adversarial attacks on IQA. We systematically evaluate 25 defense strategies, including adversarial purification, adversarial training, and certified robustness methods. We applied 14 adversarial attack algorithms of various types in both non-adaptive and adaptive settings and tested these defenses against them. We analyze the differences between defenses and their applicability to IQA tasks, considering that they should preserve IQA scores and image quality. The proposed benchmark aims to guide future developments and accepts submissions of new methods, with the latest results available online: https://videoprocessing.ai/benchmarks/iqa-defenses.html.

Eugene Lyapustin, Anastasia Kirillova, Viacheslav Meshchaninov et al.

Super-resolution (SR) has become a widely researched topic in recent years. SR methods can improve overall image and video quality and create new possibilities for further content analysis. But the SR mainstream focuses primarily on increasing the naturalness of the resulting image despite potentially losing context accuracy. Such methods may produce an incorrect digit, character, face, or other structural object even though they otherwise yield good visual quality. Incorrect detail restoration can cause errors when detecting and identifying objects both manually and automatically. To analyze the detail-restoration capabilities of image and video SR models, we developed a benchmark based on our own video dataset, which contains complex patterns that SR models generally fail to correctly restore. We assessed 32 recent SR models using our benchmark and compared their ability to preserve scene context. We also conducted a crowd-sourced comparison of restored details and developed an objective assessment metric that outperforms other quality metrics by correlation with subjective scores for this task. In conclusion, we provide a deep analysis of benchmark results that yields insights for future SR-based work.

Ekaterina Shumitskaya, Anastasia Antsiferova, Dmitriy Vatolin

No-reference image- and video-quality metrics are widely used in video processing benchmarks. The robustness of learning-based metrics under video attacks has not been widely studied. In addition to having success, attacks that can be employed in video processing benchmarks must be fast and imperceptible. This paper introduces an Invisible One-Iteration (IOI) adversarial attack on no reference image and video quality metrics. We compared our method alongside eight prior approaches using image and video datasets via objective and subjective tests. Our method exhibited superior visual quality across various attacked metric architectures while maintaining comparable attack success and speed. We made the code available on GitHub: https://github.com/katiashh/ioi-attack.

Victoria Leonenkova, Ekaterina Shumitskaya, Anastasia Antsiferova et al.

Objective no-reference image- and video-quality metrics are crucial in many computer vision tasks. However, state-of-the-art no-reference metrics have become learning-based and are vulnerable to adversarial attacks. The vulnerability of quality metrics imposes restrictions on using such metrics in quality control systems and comparing objective algorithms. Also, using vulnerable metrics as a loss for deep learning model training can mislead training to worsen visual quality. Because of that, quality metrics testing for vulnerability is a task of current interest. This paper proposes a new method for testing quality metrics vulnerability in the physical space. To our knowledge, quality metrics were not previously tested for vulnerability to this attack; they were only tested in the pixel space. We applied a physical adversarial Ti-Patch (Tiled Patch) attack to quality metrics and did experiments both in pixel and physical space. We also performed experiments on the implementation of physical adversarial wallpaper. The proposed method can be used as additional quality metrics in vulnerability evaluation, complementing traditional subjective comparison and vulnerability tests in the pixel space. We made our code and adversarial videos available on GitHub: https://github.com/leonenkova/Ti-Patch.

Victoria Leonenkova, Ekaterina Shumitskaya, Dmitriy Vatolin et al.

This demonstration presents Digital-Physical Adversarial Attacks (DiPA), a new class of practical adversarial attacks against pervasive camera-based authentication systems, where an attacker displays an adversarial patch directly on a smartphone screen instead of relying on printed artifacts. This digital-only physical presentation enables rapid deployment, removes the need for total-variation regularization, and improves patch transferability in black-box conditions. DiPA leverages an ensemble of state-of-the-art face-recognition models (ArcFace, MagFace, CosFace) to enhance transfer across unseen commercial systems. Our interactive demo shows a real-time dodging attack against a deployed face-recognition camera, preventing authorized users from being recognized while participants dynamically adjust patch patterns and observe immediate effects on the sensing pipeline. We further demonstrate DiPA's superiority over existing physical attacks in terms of success rate, feature-space distortion, and reductions in detection confidence, highlighting critical vulnerabilities at the intersection of mobile devices, pervasive vision, and sensor-driven authentication infrastructures.

Egor Kovalev, Georgii Bychkov, Khaled Abud et al.

Adversarial robustness of neural networks is an increasingly important area of research, combining studies on computer vision models, large language models (LLMs), and others. With the release of JPEG AI - the first standard for end-to-end neural image compression (NIC) methods - the question of its robustness has become critically significant. JPEG AI is among the first international, real-world applications of neural-network-based models to be embedded in consumer devices. However, research on NIC robustness has been limited to open-source codecs and a narrow range of attacks. This paper proposes a new methodology for measuring NIC robustness to adversarial attacks. We present the first large-scale evaluation of JPEG AI's robustness, comparing it with other NIC models. Our evaluation results and code are publicly available online (link is hidden for a blind review).

Nickolay Safonov, Dmitriy Vatolin

In recent years, display intensity and contrast have increased considerably. Many displays support high dynamic range (HDR) and 10-bit color depth. Since high bit-depth is an emerging technology, video content is still largely shot and transmitted with a bit depth of 8 bits or less per color component. Insufficient bit-depths produce distortions called false contours or banding, and they are visible on high contrast screens. To deal with such distortions, researchers have proposed algorithms for bit-depth enhancement (dequantization). Such techniques convert videos with low bit-depth (LBD) to videos with high bit-depth (HBD). The quality of converted LBD video, however, is usually lower than that of the original HBD video, and many consumers prefer to keep the original HBD versions. In this paper, we propose an algorithm to determine whether a video has undergone conversion before compression. This problem is complex; it involves detecting outcomes of different dequantization algorithms in the presence of compression that strongly affects the least-significant bits (LSBs) in the video frames. Our algorithm can detect bit-depth enhancement and demonstrates good generalization capability, as it is able to determine whether a video has undergone processing by dequantization algorithms absent from the training dataset.

Vladislav Bargatin, Egor Chistov, Alexander Yakovenko et al.

Recent advances in optical flow estimation have prioritized accuracy at the cost of growing GPU memory consumption, particularly for high-resolution (FullHD) inputs. We introduce MEMFOF, a memory-efficient multi-frame optical flow method that identifies a favorable trade-off between multi-frame estimation and GPU memory usage. Notably, MEMFOF requires only 2.09 GB of GPU memory at runtime for 1080p inputs, and 28.5 GB during training, which uniquely positions our method to be trained at native 1080p without the need for cropping or downsampling. We systematically revisit design choices from RAFT-like architectures, integrating reduced correlation volumes and high-resolution training protocols alongside multi-frame estimation, to achieve state-of-the-art performance across multiple benchmarks while substantially reducing memory overhead. Our method outperforms more resource-intensive alternatives in both accuracy and runtime efficiency, validating its robustness for flow estimation at high resolutions. At the time of submission, our method ranks first on the Spring benchmark with a 1-pixel (1px) outlier rate of 3.289, leads Sintel (clean) with an endpoint error (EPE) of 0.963, and achieves the best Fl-all error on KITTI-2015 at 2.94%. The code is available at https://github.com/msu-video-group/memfof.

Georgii Bychkov, Khaled Abud, Egor Kovalev et al.

Adversarial robustness of neural networks is an increasingly important area of research, combining studies on computer vision models, large language models (LLMs), and others. With the release of JPEG AI -- the first standard for end-to-end neural image compression (NIC) methods -- the question of evaluating NIC robustness has become critically significant. However, previous research has been limited to a narrow range of codecs and attacks. To address this, we present \textbf{NIC-RobustBench}, the first open-source framework to evaluate NIC robustness and adversarial defenses' efficiency, in addition to comparing Rate-Distortion (RD) performance. The framework includes the largest number of codecs among all known NIC libraries and is easily scalable. The paper demonstrates a comprehensive overview of the NIC-RobustBench framework and employs it to analyze NIC robustness. Our code is available online at https://github.com/msu-video-group/NIC-RobustBench.

Egor Chistov, Nikita Alutis, Dmitriy Vatolin

Stereoscopic videos can contain color mismatches between the left and right views due to minor variations in camera settings, lenses, and even object reflections captured from different positions. The presence of color mismatches can lead to viewer discomfort and headaches. This problem can be solved by transferring color between stereoscopic views, but traditional methods often lack quality, while neural-network-based methods can easily overfit on artificial data. The scarcity of stereoscopic videos with real-world color mismatches hinders the evaluation of different methods' performance. Therefore, we filmed a video dataset, which includes both distorted frames with color mismatches and ground-truth data, using a beam-splitter. Our second contribution is a deep multiscale neural network that solves the color-mismatch-correction task by leveraging stereo correspondences. The experimental results demonstrate the effectiveness of the proposed method on a conventional dataset, but there remains room for improvement on challenging real-world data.

Maksim Siniukov, Dmitriy Kulikov, Dmitriy Vatolin

Subjective image-quality measurement plays a critical role in the development of image-processing applications. The purpose of a visual-quality metric is to approximate the results of subjective assessment. In this regard, more and more metrics are under development, but little research has considered their limitations. This paper addresses that deficiency: we show how image preprocessing before compression can artificially increase the quality scores provided by the popular metrics DISTS, LPIPS, HaarPSI, and VIF as well as how these scores are inconsistent with subjective-quality scores. We propose a series of neural-network preprocessing models that increase DISTS by up to 34.5%, LPIPS by up to 36.8%, VIF by up to 98.0%, and HaarPSI by up to 22.6% in the case of JPEG-compressed images. A subjective comparison of preprocessed images showed that for most of the metrics we examined, visual quality drops or stays unchanged, limiting the applicability of these metrics.

Andrey Moskalenko, Mikhail Erofeev, Dmitriy Vatolin

In recent years, the field of image inpainting has developed rapidly, learning based approaches show impressive results in the task of filling missing parts in an image. But most deep methods are strongly tied to the resolution of the images on which they were trained. A slight resolution increase leads to serious artifacts and unsatisfactory filling quality. These methods are therefore unsuitable for interactive image processing. In this article, we propose a method that solves the problem of inpainting arbitrary-size images. We also describe a way to better restore texture fragments in the filled area. For this, we propose to use information from neighboring pixels by shifting the original image in four directions. Moreover, this approach can work with existing inpainting models, making them almost resolution independent without the need for retraining. We also created a GIMP plugin that implements our technique. The plugin, code, and model weights are available at https://github.com/a-mos/High_Resolution_Image_Inpainting.

Daria Tsereh, Mark Mirgaleev, Ivan Molodetskikh et al.

Learning-based image compression methods have improved in recent years and started to outperform traditional codecs. However, neural-network approaches can unexpectedly introduce visual artifacts in some images. We therefore propose methods to separately detect three types of artifacts (texture and boundary degradation, color change, and text corruption), to localize the affected regions, and to quantify the artifact strength. We consider only those regions that exhibit distortion due solely to the neural compression but that a traditional codec recovers successfully at a comparable bitrate. We employed our methods to collect artifacts for the JPEG AI verification model with respect to HM-18.0, the H.265 reference software. We processed about 350,000 unique images from the Open Images dataset using different compression-quality parameters; the result is a dataset of 46,440 artifacts validated through crowd-sourced subjective assessment. Our proposed dataset and methods are valuable for testing neural-network-based image codecs, identifying bugs in these codecs, and enhancing their performance. We make source code of the methods and the dataset publicly available.

Aleksandr Gushchin, Anna Chistyakova, Vladislav Minashkin et al.

Recently, the area of adversarial attacks on image quality metrics has begun to be explored, whereas the area of defences remains under-researched. In this study, we aim to cover that case and check the transferability of adversarial purification defences from image classifiers to IQA methods. In this paper, we apply several widespread attacks on IQA models and examine the success of the defences against them. The purification methodologies covered different preprocessing techniques, including geometrical transformations, compression, denoising, and modern neural network-based methods. Also, we address the challenge of assessing the efficacy of a defensive methodology by proposing ways to estimate output visual quality and the success of neutralizing attacks. Defences were tested against attack on three IQA metrics -- Linearity, MetaIQA and SPAQ. The code for attacks and defences is available at: (link is hidden for a blind review).

Mikhail Dremin, Konstantin Kozhemyakov, Ivan Molodetskikh et al.

A main goal in developing video-compression algorithms is to enhance human-perceived visual quality while maintaining file size. But modern video-analysis efforts such as detection and recognition, which are integral to video surveillance and autonomous vehicles, involve so much data that they necessitate machine-vision processing with minimal human intervention. In such cases, the video codec must be optimized for machine vision. This paper explores the effects of compression on detection and recognition algorithms (objects, faces, and license plates) and introduces novel full-reference image/video-quality metrics for each task, tailored to machine vision. Experimental results indicate our proposed metrics correlate better with the machine-vision results for the respective tasks than do existing image/video-quality metrics.

Kirill Koltsov, Aleksandr Gushchin, Dmitriy Vatolin et al.

Text rendering remains a persistent failure mode of modern text-to-image models (T2I), yet existing evaluations rely on OCR correctness or VLM-based judging procedures that are poorly aligned with perceptual text artifacts. We introduce Text-in-Image Quality Assessment (TIQA), a task that predicts a scalar quality score that matches human judgments of rendered-text fidelity within cropped text regions. We release two MOS-labeled datasets: TIQA-Crops (10k text crops) and TIQA-Images (1,500 images), spanning 20+ T2I models, including proprietary ones. We also propose ANTIQA, a lightweight method with text-specific biases, and show that it improves correlation with human scores over OCR confidence, VLM judges, and generic NR-IQA metrics by at least $\sim0.05$ on TIQA-Crops and $\sim0.08$ on TIQA-Images, as measured by PLCC. Finally, we show that TIQA models are valuable in downstream tasks: for example, selecting the best-of-5 generations with ANTIQA improves human-rated text quality by $+14\%$ on average, demonstrating practical value for filtering and reranking in generation pipelines.

Ivan Molodetskikh, Kirill Malyshev, Mark Mirgaleev et al.

Generative image super-resolution (SR) is rapidly advancing in visual quality and detail restoration. As the capacity of SR models expands, however, so does their tendency to produce artifacts: incorrect, visually disturbing details that reduce perceived quality. Crucially, their perceptual impact varies: some artifacts are barely noticeable while others strongly degrade the image. We argue that artifacts should be characterized by their prominence to human observers rather than treated as uniform binary defects. Motivated by this, we present a novel dataset of 1302 artifact examples from 11 contemporary image-SR methods, where each artifact is paired with a crowdsourced prominence score. Building on this dataset, we train a lightweight regressor that produces spatial prominence heatmaps and outperforms existing methods at detecting prominent artifacts. We release the dataset and code to facilitate prominence-aware evaluation and mitigation of SR artifacts.

Ekaterina Shumitskaya, Dmitriy Vatolin, Anastasia Antsiferova

We propose a novel certified defense method for Image Quality Assessment (IQA) models based on randomized smoothing with noise applied in the feature space rather than the input space. Unlike prior approaches that inject Gaussian noise directly into input images, often degrading visual quality, our method preserves image fidelity while providing robustness guarantees. To formally connect noise levels in the feature space with corresponding input-space perturbations, we analyze the maximum singular value of the backbone network's Jacobian. Our approach supports both full-reference (FR) and no-reference (NR) IQA models without requiring any architectural modifications, suitable for various scenarios. It is also computationally efficient, requiring a single backbone forward pass per image. Compared to previous methods, it reduces inference time by 99.5% without certification and by 20.6% when certification is applied. We validate our method with extensive experiments on two benchmark datasets, involving six widely-used FR and NR IQA models and comparisons against five state-of-the-art certified defenses. Our results demonstrate consistent improvements in correlation with subjective quality scores by up to 30.9%.

Aleksandr Gushchin, Maksim Smirnov, Dmitriy Vatolin et al.

We propose the LEHA-CVQAD (Large-scale Enriched Human-Annotated Compressed Video Quality Assessment) dataset, which comprises 6,240 clips for compression-oriented video quality assessment. 59 source videos are encoded with 186 codec-preset variants, 1.8M pairwise, and 1.5k MOS ratings are fused into a single quality scale; part of the videos remains hidden for blind evaluation. We also propose Rate-Distortion Alignment Error (RDAE), a novel evaluation metric that quantifies how well VQA models preserve bitrate-quality ordering, directly supporting codec parameter tuning. Testing IQA/VQA methods reveals that popular VQA metrics exhibit high RDAE and lower correlations, underscoring the dataset challenges and utility. The open part and the results of LEHA-CVQAD are available at https://aleksandrgushchin.github.io/lcvqad/

Igor Meleshin, Anna Chistyakova, Anastasia Antsiferova et al.

Image Quality Assessment (IQA) models are increasingly relied upon to evaluate image quality in real-world systems -- from compression and enhancement to generation and streaming. Yet their adoption brings a fundamental risk: these models are inherently unstable. Adversarial manipulations can easily fool them, inflating scores and undermining trust. Traditionally, such vulnerabilities are addressed through data-driven defenses -- adversarial retraining, regularization, or input purification. But what if this is the wrong lens? What if robustness in perceptual models is not something to learn but something to design? In this work, we propose a provocative idea: robustness as an architectural prior. Rather than training models to resist perturbations, we reshape their internal structure to suppress sensitivity from the ground up. We achieve this by enforcing orthogonal information flow, constraining the network to norm-preserving operations -- and further stabilizing the system through pruning and fine-tuning. The result is a robust IQA architecture that withstands adversarial attacks without requiring adversarial training or significant changes to the original model. This approach suggests a shift in perspective: from optimizing robustness through data to engineering it through design.

Georgii Gotin, Ekaterina Shumitskaya, Anastasia Antsiferova et al.

Recent studies have revealed that modern image and video quality assessment (IQA/VQA) metrics are vulnerable to adversarial attacks. An attacker can manipulate a video through preprocessing to artificially increase its quality score according to a certain metric, despite no actual improvement in visual quality. Most of the attacks studied in the literature are white-box attacks, while black-box attacks in the context of VQA have received less attention. Moreover, some research indicates a lack of transferability of adversarial examples generated for one model to another when applied to VQA. In this paper, we propose a cross-modal attack method, IC2VQA, aimed at exploring the vulnerabilities of modern VQA models. This approach is motivated by the observation that the low-level feature spaces of images and videos are similar. We investigate the transferability of adversarial perturbations across different modalities; specifically, we analyze how adversarial perturbations generated on a white-box IQA model with an additional CLIP module can effectively target a VQA model. The addition of the CLIP module serves as a valuable aid in increasing transferability, as the CLIP model is known for its effective capture of low-level semantics. Extensive experiments demonstrate that IC2VQA achieves a high success rate in attacking three black-box VQA models. We compare our method with existing black-box attack strategies, highlighting its superiority in terms of attack success within the same number of iterations and levels of attack strength. We believe that the proposed method will contribute to the deeper analysis of robust VQA metrics.

Khaled Abud, Sergey Lavrushkin, Alexey Kirillov et al.

Diffusion-based models have recently revolutionized image generation, achieving unprecedented levels of fidelity. However, consistent generation of high-quality images remains challenging partly due to the lack of conditioning mechanisms for perceptual quality. In this work, we propose methods to integrate image quality assessment (IQA) models into diffusion-based generators, enabling quality-aware image generation. We show that diffusion models can learn complex qualitative relationships from both IQA models' outputs and internal activations. First, we experiment with gradient-based guidance to optimize image quality directly and show this method has limited generalizability. To address this, we introduce IQA-Adapter, a novel framework that conditions generation on target quality levels by learning the implicit relationship between images and quality scores. When conditioned on high target quality, IQA-Adapter can shift the distribution of generated images towards a higher-quality subdomain, and, inversely, it can be used as a degradation model, generating progressively more distorted images when provided with a lower-quality signal. Under high-quality condition, IQA-Adapter achieves up to a 10% improvement across multiple objective metrics, as confirmed by a user preference study, while preserving generative diversity and content. Furthermore, we extend IQA-Adapter to a reference-based conditioning scenario, utilizing the rich activation space of IQA models to transfer highly specific, content-agnostic qualitative features between images.

Ekaterina Shumitskaya, Mikhail Pautov, Dmitriy Vatolin et al.

Most modern No-Reference Image-Quality Assessment (NR-IQA) metrics are based on neural networks vulnerable to adversarial attacks. Attacks on such metrics lead to incorrect image/video quality predictions, which poses significant risks, especially in public benchmarks. Developers of image processing algorithms may unfairly increase the score of a target IQA metric without improving the actual quality of the adversarial image. Although some empirical defenses for IQA metrics were proposed, they do not provide theoretical guarantees and may be vulnerable to adaptive attacks. This work focuses on developing a provably robust no-reference IQA metric. Our method is based on Median Smoothing (MS) combined with an additional convolution denoiser with ranking loss to improve the SROCC and PLCC scores of the defended IQA metric. Compared with two prior methods on three datasets, our method exhibited superior SROCC and PLCC scores while maintaining comparable certified guarantees.

Kirillov Alexey, Andrey Moskalenko, Dmitriy Vatolin

Over the past few years, deep neural models have made considerable advances in image quality assessment (IQA). However, the underlying reasons for their success remain unclear, owing to the complex nature of deep neural networks. IQA aims to describe how the human visual system (HVS) works and to create its efficient approximations. On the other hand, Saliency Prediction task aims to emulate HVS via determining areas of visual interest. Thus, we believe that saliency plays a crucial role in human perception. In this work, we conduct an empirical study that reveals the relation between IQA and Saliency Prediction tasks, demonstrating that the former incorporates knowledge of the latter. Moreover, we introduce a novel SACID dataset of saliency-aware compressed images and conduct a large-scale comparison of classic and neural-based IQA methods. All supplementary code and data will be available at the time of publication.

Nikita Alutis, Egor Chistov, Mikhail Dremin et al.

This paper discusses the challenges of evaluating deblurring-methods quality and proposes a reduced-reference metric based on machine learning. Traditional quality-assessment metrics such as PSNR and SSIM are common for this task, but not only do they correlate poorly with subjective assessments, they also require ground-truth (GT) frames, which can be difficult to obtain in the case of deblurring. To develop and evaluate our metric, we created a new motion-blur dataset using a beam splitter. The setup captured various motion types using a static camera, as most scenes in existing datasets include blur due to camera motion. We also conducted two large subjective comparisons to aid in metric development. Our resulting metric requires no GT frames, and it correlates well with subjective human perception of blur.

Ekaterina Shumitskaya, Anastasia Antsiferova, Dmitriy Vatolin

Modern neural-network-based no-reference image- and video-quality metrics exhibit performance as high as full-reference metrics. These metrics are widely used to improve visual quality in computer vision methods and compare video processing methods. However, these metrics are not stable to traditional adversarial attacks, which can cause incorrect results. Our goal is to investigate the boundaries of no-reference metrics applicability, and in this paper, we propose a fast adversarial perturbation attack on no-reference quality metrics. The proposed attack (FACPA) can be exploited as a preprocessing step in real-time video processing and compression algorithms. This research can yield insights to further aid in designing of stable neural-network-based no-reference quality metrics.

Evgeney Bogatyrev, Ivan Molodetskikh, Dmitriy Vatolin

In recent years, there has been significant interest in Super-Resolution (SR), which focuses on generating a high-resolution image from a low-resolution input. Deep learning-based methods for super-resolution have been particularly popular and have shown impressive results on various benchmarks. However, research indicates that these methods may not perform as well on strongly compressed videos. We developed a super-resolution benchmark to analyze SR's capacity to upscale compressed videos. Our dataset employed video codecs based on five widely-used compression standards: H.264, H.265, H.266, AV1, and AVS3. We assessed 19 popular SR models using our benchmark and evaluated their ability to restore details and their susceptibility to compression artifacts. To get an accurate perceptual ranking of SR models, we conducted a crowd-sourced side-by-side comparison of their outputs. We found that some SR models, combined with compression, allow us to reduce the video bitrate without significant loss of quality. We also compared a range of image and video quality metrics with subjective scores to evaluate their accuracy on super-resolved compressed videos. The benchmark is publicly available at https://videoprocessing.ai/benchmarks/super-resolution-for-video-compression.html

Roman Kazantsev, Dmitriy Vatolin

The critical constraint of mobile devices is a limited battery life that is significantly reduced during video playback. The power efficiency of video playback mainly depends on the used compression standard, video-decoder, and device model. We propose a software-based method to estimate the power consumption of video-decoders on various Android devices. Experiments on two devices of the same model show a small variation of the power playback consumption and a lack of dependence between the power consumption and the battery level. We have implemented an automatic system that includes the VEQE Android application to measure the power consumption of decoders and a server to collect the power metrics. Our system has collected power-consumption and decoding-speed dataset for video-decoders of six standards (AV1, HEVC, VP9, H.264, VP8, and MPEG-4) operating on 285 devices, representing 147 models. We demonstrate some slices of the created dataset: the top 30 models and video-decoders in terms of power efficiency for playback and for decoding only, as well as video-decoder ratings by power consumption and decoding speed for a given device model.

Alexander Gushchin, Anastasia Antsiferova, Dmitriy Vatolin

Shot boundary detection in video is one of the key stages of video data processing. A new method for shot boundary detection based on several video features, such as color histograms and object boundaries, has been proposed. The developed algorithm was tested on the open BBC Planet Earth [1] and RAI [2] datasets, and the MSU CC datasets, based on videos used in the video codec comparison conducted at MSU, as well as videos from the IBM set, were also plotted. The total dataset for algorithm development and testing exceeded the known TRECVID datasets. Based on the test results, the proposed algorithm for scene change detection outperformed its counterparts with a final F-score of 0.9794.

Anastasia Antsiferova, Alexander Yakovenko, Nickolay Safonov et al.

Quality assessment plays a key role in creating and comparing video compression algorithms. Despite the development of a large number of new methods for assessing quality, generally accepted and well-known codecs comparisons mainly use the classical methods like PSNR, SSIM and new method VMAF. These methods can be calculated following different rules: they can use different frame-by-frame averaging techniques or different summation of color components. In this paper, a fundamental comparison of various versions of generally accepted metrics is carried out to find the most relevant and recommended versions of video quality metrics to be used in codecs comparisons. For comparison, we used a set of videos encoded with video codecs of different standards, and visual quality scores collected for the resulting set of streams since 2018 until 2021

Maksim Siniukov, Anastasia Antsiferova, Dmitriy Kulikov et al.

Video-quality measurement plays a critical role in the development of video-processing applications. In this paper, we show how video preprocessing can artificially increase the popular quality metric VMAF and its tuning-resistant version, VMAF NEG. We propose a pipeline that tunes processing-algorithm parameters to increase VMAF by up to 218.8%. A subjective comparison revealed that for most preprocessing methods, a video's visual quality drops or stays unchanged. We also show that some preprocessing methods can increase VMAF NEG scores by up to 23.6%.

Anastasia Zvezdakova, Sergey Zvezdakov, Dmitriy Kulikov et al.

Video quality measurement takes an important role in many applications. Full-reference quality metrics which are usually used in video codecs comparisons are expected to reflect any changes in videos. In this article, we consider different color corrections of compressed videos which increase the values of full-reference metric VMAF and almost don't decrease other widely-used metric SSIM. The proposed video contrast enhancement approach shows the metric inapplicability in some cases for video codecs comparisons, as it may be used for cheating in the comparisons via tuning to improve this metric values.

Anastasia Zvezdakova, Dmitriy Kulikov, Denis Kondranin et al.

This paper analyses the application of no-reference metric NIQE to the task of video-codec comparison. A number of issues in the metric behaviour on videos was detected and described. The metric has outlying scores on black and solid-coloured frames. The proposed averaging technique for metric quality scores helped to improve the results in some cases. Also, NIQE has low-quality scores for videos with detailed textures and higher scores for videos of lower bitrates due to the blurring of these textures after compression. Although NIQE showed natural results for many tested videos, it is not universal and currently can not be used for video-codec comparisons.

Vitaliy Lyudvichenko, Dmitriy Vatolin

This paper presents a new way of getting high-quality saliency maps for video, using a cheaper alternative to eye-tracking data. We designed a mouse-contingent video viewing system which simulates the viewers' peripheral vision based on the position of the mouse cursor. The system enables the use of mouse-tracking data recorded from an ordinary computer mouse as an alternative to real gaze fixations recorded by a more expensive eye-tracker. We developed a crowdsourcing system that enables the collection of such mouse-tracking data at large scale. Using the collected mouse-tracking data we showed that it can serve as an approximation of eye-tracking data. Moreover, trying to increase the efficiency of collected mouse-tracking data we proposed a novel deep neural network algorithm that improves the quality of mouse-tracking saliency maps.

Vitaliy Lyudvichenko, Mikhail Erofeev, Alexander Ploshkin et al.

Recent advances in deep learning have markedly improved the quality of visual-attention modelling. In this work we apply these advances to video compression. We propose a compression method that uses a saliency model to adaptively compress frame areas in accordance with their predicted saliency. We selected three state-of-the-art saliency models, adapted them for video compression and analyzed their results. The analysis includes objective evaluation of the models as well as objective and subjective evaluation of the compressed videos. Our method, which is based on the x264 video codec, can produce videos with the same visual quality as regular x264, but it reduces the bitrate by 25% according to the objective evaluation and by 17% according to the subjective one. Also, both the subjective and objective evaluations demonstrate that saliency models can compete with gaze maps for a single observer. Our method can extend to most video bitstream formats and can improve video compression quality without requiring a switch to a new video encoding standard.