Sergey Lavrushkin has a total of 38 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Investigating the Robustness of Multi-Model Driving Perception to a Joint Physical Patch

    by Sergey Muravlev, Sergey Lavrushkin, Daniel Kanta, Georgii Bychkov, Dmitriy Vatolin

  • Open Access

    VCR-Bench: A Modular Open-Source Benchmark for Video Classification Robustness

    by Plinskiy Maksim, Aleksandr Gushchin, Sergey Lavrushkin, Dmitriy S. Vatolin, Anastasia Antsiferova

  • Open Access

    WARP: A Unified Benchmark for Invisible Image Watermarking - Robustness and Protection Against Attacks

    by Khaled Abud, Aleksey Yakushev, Aleksandr Akimenkov, Irina Serzhenko, Kirill Aistov, Egor Kovalev, Dmitry Obydenkov, Sergey Lavrushkin, Anastasia Antsiferova, Dmitriy Vatolin, et al.

  • Open Access

    A software system for automated assessment of stereoscopic distortions in VR180 videos

    by Sergey V. Lavrushkin

    The paper presents a comprehensive software system for automatically evaluating stereoscopic distortions in VR180 video. The proposed approach takes into account the most common types of artifacts: color mismatches, differences in sharpness, geometric distortions (vertical shift, rotation, scaling), and channel mismatches. Specialized algorithms have been developed for each type of distortion, based on disparity maps, motion vectors and their confidence maps, as well as neural network regression or classification methods. The proposed solutions have been successfully tested on several datasets, demonstrating high accuracy in detecting various types of distortions in VR180 video. The system can be integrated into standard post-processing pipelines and provides automated generation of detailed reports, allowing stereoscopic content creators to quickly identify and eliminate stereoscopic distortions before releasing their products to a wide audience. В статье представлена комплексная программная система для автоматической оценки стереоскопических искажений видео в формате VR180. Предложенный подход учитывает наиболее распространённые типы артефактов: несоответствия по цвету, различия по резкости, геометрические искажения (вертикальный сдвиг, поворот, масштабирование), а также перепутанный порядок ракурсов. Для каждого из перечисленных искажений разработаны специализированные алгоритмы, основанные на вычислении карт диспаратности, векторов движения, карт доверия к ним и нейросетевых методов регрессии или классификации. Предложенные решения успешно прошли проверку на нескольких тестовых наборах данных и продемонстрировали высокую точность в обнаружении искажений разных типов для VR180-видео. Система может быть интегрирована в стандартные пайплайны постобработки и обеспечивает автоматизированную генерацию подробных отчётов, что может позволить создателям стереоконтента оперативно выявлять и устранять причины стереоскопических искажений до вывода продукции на массового зрителя.

  • Open Access

    Evaluating Adversarial Robustness of No-Reference Image and Video Quality Assessment Models with Frequency-Masked Gradient Orthogonalization Adversarial Attack

    by Khaled Abud, Sergey Lavrushkin, Dmitry Vatolin

    Neural-network-based models have made considerable progress in many computer vision areas over recent years. However, many works have exposed their vulnerability to malicious input data manipulation—that is, to adversarial attacks. Although many recent works have thoroughly examined the adversarial robustness of classifiers, the robustness of Image Quality Assessment (IQA) methods remains understudied. This paper addresses this gap by proposing FM-GOAT (Frequency-Masked Gradient Orthogonalization Attack), a novel white box adversarial method tailored for no-reference IQA models. Using a novel gradient orthogonalization technique, FM-GOAT uniquely optimizes adversarial perturbations against multiple perceptual constraints to minimize visibility, moving beyond traditional lp-norm bounds. We evaluate FM-GOAT on seven state-of-the-art NR-IQA models across three image and video datasets, revealing significant vulnerability to the proposed attack. Furthermore, we examine the applicability of adversarial purification methods to the IQA task, as well as their efficiency in mitigating white box adversarial attacks. By studying the activations from models’ intermediate layers, we explore their behavioral patterns in adversarial scenarios and discover valuable insights that may lead to better adversarial detection.


  • Stable VMAF: investigating VMAF’s vulnerabilities to adversarial attacks

    by Sergey Lavrushkin, Maksim Khrebtov, Anastasia Antsiferova, Georgii Bychkov, Alexey Soloviev, Dmitriy Vatolin

  • Open Access

    Exploring adversarial robustness of JPEG AI: methodology, comparison and new methods

    by Egor Kovalev, Georgii Bychkov, Khaled Abud, Aleksandr Gushchin, Anna Chistyakova, Sergey Lavrushkin, Dmitriy Vatolin, Anastasia Antsiferova

    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).

  • Open Access

    Increasing the Robustness of Image Quality Assessment Models Through Adversarial Training

    by Anna Chistyakova, Anastasia Antsiferova, Maksim Khrebtov, Sergey Lavrushkin, Konstantin Arkhipenko, Dmitriy Vatolin, Denis Turdakov

    The adversarial robustness of image quality assessment (IQA) models to adversarial attacks is emerging as a critical issue. Adversarial training has been widely used to improve the robustness of neural networks to adversarial attacks, but little in-depth research has examined adversarial training as a way to improve IQA model robustness. This study introduces an enhanced adversarial training approach tailored to IQA models; it adjusts the perceptual quality scores of adversarial images during training to enhance the correlation between an IQA model’s quality and the subjective quality scores. We also propose a new method for comparing IQA model robustness by measuring the Integral Robustness Score; this method evaluates the IQA model resistance to a set of adversarial perturbations with different magnitudes. We used our adversarial training approach to increase the robustness of five IQA models. Additionally, we tested the robustness of adversarially trained IQA models to 16 adversarial attacks and conducted an empirical probabilistic estimation of this feature.


  • Open Access

    Increasing the Robustness of Image Quality Assessment Models Through Adversarial Training

    by Anna Chistyakova, Anastasia Antsiferova, Maksim Khrebtov, Sergey Lavrushkin, Konstantin Arkhipenko, Dmitriy Vatolin, Denis Turdakov

    The adversarial robustness of image quality assessment (IQA) models to adversarial attacks is emerging as a critical issue. Adversarial training has been widely used to improve the robustness of neural networks to adversarial attacks, but little in-depth research has examined adversarial training as a way to improve IQA model robustness. This study introduces an enhanced adversarial training approach tailored to IQA models; it adjusts the perceptual quality scores of adversarial images during training to enhance the correlation between an IQA model’s quality and the subjective quality scores. We also propose a new method for comparing IQA model robustness by measuring the Integral Robustness Score; this method evaluates the IQA model resistance to a set of adversarial perturbations with different magnitudes. We used our adversarial training approach to increase the robustness of five IQA models. Additionally, we tested the robustness of adversarially trained IQA models to 16 adversarial attacks and conducted an empirical probabilistic estimation of this feature. The code is available at https://github.com/wianluna/metrics_at.


  • Open Access

    Comparing the Robustness of Modern No-Reference Image- and Video-Quality Metrics to Adversarial Attacks

    by Anastasia Antsiferova, Anastasia Antsiferova, Anastasia Antsiferova, Khaled Abud, Khaled Abud, Khaled Abud, Aleksandr Gushchin, Aleksandr Gushchin, Aleksandr Gushchin, Ekaterina Shumitskaya, 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.