PNPanos Nasiopoulos

Panos Nasiopoulos

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Panos Nasiopoulos has a total of 51 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • An Innovative Perceptual Wavelet Packet-Based Flicker Metric for VRR Displays

    by Hamid Reza Tohidypour, Frank Seto, Panos Nasiopoulos

    Variable refresh rate (VRR) technology has become increasingly popular in consumer displays, offering benefits such as reduced input lag, minimized video stuttering, and improved energy efficiency. However, unlike fixed-refresh-rate displays, VRR screens refresh at varying intervals, leading to subtle luminance fluctuations that can cause perceptible flicker, negatively affecting user comfort and viewing satisfaction. Existing flicker assessment methods are either overly restrictive or rely on resource-intensive subjective evaluations, limiting their practical applicability across different display types. To address this challenge, we introduce a novel wavelet packet-based VRR flicker metric that analyzes and perceptually weights the luminance signal based on the temporal contrast sensitivity function (TCSF) to provide an accurate, objective flicker measurement. Our approach identifies the Stela TCSF as the most effective for wavelet based VRR flicker evaluation and demonstrates that the Daubechies 2 wavelet performs best in capturing relevant flicker characteristics. We validated our metric through a comprehensive subjective study involving diverse VRR sequences with varying luminance levels and display technologies. The proposed metric achieved a Pearson correlation of 90.32% with mean opinion scores (MOS), outperforming the leading industry metric by 18.89%. Our method offers a reliable and efficient solution for quantifying flicker in VRR displays, enabling manufacturers to enhance display performance and provide consumers with smoother and more comfortable viewing experiences.

  • Open Access

    Shallow- and Deep-fake Image Manipulation Localization Using Vision Mamba and Guided Graph Neural Network

    by Junbin Zhang, Hamid Reza Tohidypour, Yixiao Wang, Panos Nasiopoulos

    Image manipulation localization is a critical research task, given that forged images may have a significant societal impact of various aspects. Such image manipulations can be produced using traditional image editing tools (known as "shallowfakes") or advanced artificial intelligence techniques ("deepfakes"). While numerous studies have focused on image manipulation localization on either shallowfake images or deepfake videos, few approaches address both cases. In this paper, we explore the feasibility of using a deep learning network to localize manipulations in both shallow- and deep-fake images, and proposed a solution for such purpose. To precisely differentiate between authentic and manipulated pixels, we leverage the Vision Mamba network to extract feature maps that clearly describe the boundaries between tampered and untouched regions. To further enhance this separation, we propose a novel Guided Graph Neural Network (G-GNN) module that amplifies the distinction between manipulated and authentic pixels. Our evaluation results show that our proposed method achieved higher inference accuracy compared to other state-of-the-art methods.


  • Video Compression Artifacts Reduction with Recurrent Video Restoration Transformer

    by Zhenchao Ma, Berkay Talha Acar, Danni Zhao, Maya Thomas, Hamid Reza Tohidypour, Panos Nasiopoulos

  • StereoMamba+: A Novel Stereo Image Super-Resolution Framework With Adaptive Dependency Capture and Enhanced Feature Fusion

    by Zhenchao Ma, Hamid Reza Tohidypour, Panos Nasiopoulos, Victor C. M. Leung

    Stereo image Super-Resolution (SR) aims to enhance image resolution by leveraging complementary information in stereo pairs. Convolutional Neural Networks (CNNs), widely used in stereo image SR for their strong local pattern extraction capabilities, often fail to capture long-range dependencies critical for stereo correspondence. On the other hand, Swin Transformers have demonstrated superior performance in modeling long-range dependencies for stereo image SR tasks. However, their computational complexity scales quadratically with the window size, leading to a trade-off between global receptive fields and computational efficiency. To tackle these challenges, we propose StereoMamba+, a novel stereo image SR method designed to adaptively capture both local and global dependencies in stereo pairs. Leveraging the Mamba architecture as its backbone, StereoMamba+ integrates an Adaptive State Space Module (ASSM) that efficiently extracts and fuses global and local features, maintaining linear computational complexity. Additionally, a Gated Enhanced Feed-Forward Network (GEFN) selectively amplifies essential features and depth cues, and a Residual Frequency Block (RFB) is employed to capture global features in the frequency domain. To further enhance stereo correspondence, we introduce a Stereo Bi-Directional Cross Attention Module (SBCAM), aligning unique features along both horizontal and vertical epipolar lines to improve stereo consistency. Extensive experiments demonstrate that our proposed StereoMamba+ method achieves state-of-the-art performance on 2× and 4× stereo image SR tasks, delivering PSNR improvements of up to 0.45dB, while maintaining competitive parameter efficiency compared to existing methods.


  • A Novel Image Enhancement Method for Dimmed Displays Based on JND and Human Frequency Sensitivity

    by Hamid Reza Tohidypour, Panos Nasiopoulos, Kristyn Lee, Catherina Liu, George Du, Yubo Wang

  • Leveraging Bayer Pattern Analysis for Authenticity Detection of Real and Fake Images

    by Junbin Zhang, Hamid Reza Tohidypour, Panos Nasiopoulos, Larry Liu, Ziyao Wang, Eric Cao, Olasunade Ladenegan


  • Open Access

    An Efficient Frequency Domain Based Attribution and Detection Network

    by Junbin Zhang, Yixiao Wang, Hamid Reza Tohidypour, Panos Nasiopoulos

    People nowadays can easily synthesize high fidelity fake images with different types of image content due to the rapid advances of deep learning technologies. Detecting such images and attributing them to their generative models (GMs) is crucial. Existing deep learning methods attempt to identify and classify GM-specific artifacts but often struggle with content-independence and generalizability. In this paper, we observe that while GMs leave unique artifacts in the frequency domain, they are coupled with the image content. Based on this observation, we propose a novel deep learning-based solution that learns input-adaptive masks to highlight GMs’ artifacts and achieve high accuracy on the synthesized image attribution task. In addition, we observed that GMs’ artifacts in the frequency domain remain intact in sub-images of the original image, and they are even retained when the images are distorted. To further improve the accuracy of the proposed solution, we leverage the characteristics of GMs artifacts in sub-images and distorted images to make our network perform more effectively. Our evaluation results show that our proposed solution outperforms other state-of-the-art methods on unseen image types, showing great generalizability.


  • Accurate Detection and Localization of Individual Free Street Parking Spaces Using AI and Innovative Global Motion Estimation

    by Tala Bazzaza, Hamid Reza Tohidypour, Yixiao Wang, Panos Nasiopoulos

    Traffic congestion has become a universal phenomenon in metropolitan cities. The prevalence of congested roads not only results in delays, increased fuel consumption, environmental pollution, and a reduced quality of life, but also contributes to lowered productivity due to the time wasted in the search for available street parking. Current solutions are either tailored solely for detecting vacant spaces within parking lots or rely on unfeasible street sensors and stationary surveillance cameras, falling short in accurately pinpointing individual free street parking spots. In this paper, we introduce an innovative street parking detection and localization system designed to cater to both human-driven and autonomous vehicles. Our solution is specifically engineered to be integrated into modern vehicles, utilizing the live video feed from the car's built-in navigation and obstacle avoidance cameras. This approach combines convolutional neural networks, our video global motion analysis, and an innovative distance measurement technique to achieve accurate detection and precise localization of unoccupied street parking spaces. Our solution eliminates the impact of vehicle speed, duplicate detection of identical parking spots during sudden stops, and delivers outstanding accuracy while maintaining low computational costs and complexity. We also introduce a unique dataset tailored for this objective, serving as the cornerstone for training and assessing well established object detection network architectures. Performance evaluations confirm the proposed method's efficacy across all types of scenarios


  • A Deep Learning Based Automatic Outdoor Home Video Surveillance Approach

    by Hamid Reza Tohidypour, Tala Bazzaza, Yixiao Wang, Panos Nasiopoulos, Vincent Sastra, Bowei Ren, Elbert Ng, Sebastian Gonzalez, Andrew Shieh

  • Enhancing Image Quality by Reducing Compression Artifacts Using Dynamic Window Swin Transformer

    by Zhenchao Ma, Yixiao Wang, Hamid Reza Tohidypour, Panos Nasiopoulos, Victor C. M. Leung

    Video/image compression codecs utilize the characteristics of the human visual system and its varying sensitivity to certain frequencies, brightness, contrast, and colors to achieve high compression. Inevitably, compression introduces undesirable visual artifacts. As compression standards improve, restoring image quality becomes more challenging. Recently, deep learning based models, especially transformer-based image restoration models, have emerged as a promising approach for reducing compression artifacts, demonstrating very good restoration performance. However, all the proposed transformer based restoration methods use a same fixed window size, confining pixel dependencies in fixed areas. In this paper, we propose a new and unique image restoration method that addresses the shortcoming of existing methods by first introducing a content adaptive dynamic window that is applied to self-attention layers which in turn are weighted by our channel and spatial attention module utilized in Swin Transformer to mainly capture long and medium range pixel dependencies. In addition, local dependencies are further enhanced by integrating a CNN based network inside the Swin Transformer Block to process the image augmented by our self-attention module. Performance evaluations using images compressed by one of the latest compression standards, namely the Versatile Video Coding (VVC), when measured in Peak Signal-to-Noise Ratio (PSNR), our proposed approach achieves an average gain of 1.32dB on three different benchmark datasets for VVC compression artifacts reduction. Additionally, our proposed approach improves the visual quality of compressed images by an average of 2.7% in terms of Video Multimethod Assessment Fusion (VMAF).