This paper reports on the NTIRE 2023 Quality Assessment of Video Enhancement Challenge, which will be held in conjunction with the New Trends in Image Restoration and Enhancement Workshop (NTIRE) at CVPR 2023. This challenge is to address a major challenge in the field of video processing, namely, video quality assessment (VQA) for enhanced videos. The challenge uses the VQA Dataset for Perceptual Video Enhancement (VDPVE), which has a total of 1211 enhanced videos, including 600 videos with color, brightness, and contrast enhancements, 310 videos with deblurring, and 301 deshaked videos. The challenge has a total of 167 registered participants. 61 participating teams submitted their prediction results during the development phase, with a total of 3168 submissions. A total of 176 submissions were submitted by 37 participating teams during the final testing phase. Finally, 19 participating teams submitted their models and fact sheets, and detailed the methods they used. Some methods have achieved better results than baseline methods, and the winning methods have demonstrated superior prediction performance.
Over the last few years, neural image compression has gained wide attention from research and industry, yielding promising end-to-end deep neural codecs outperforming their conventional counterparts in rate-distortion performance. Despite significant advancement, current methods, including attention-based transform coding, still need to be improved in reducing the coding rate while preserving the reconstruction fidelity, especially in non-homogeneous textured image areas. Those models also require more parameters and a higher decoding time. To tackle the above challenges, we propose ConvNeXt-ChARM, an efficient ConvNeXt-based transform coding framework, paired with a compute-efficient channel-wise auto-regressive prior to capturing both global and local contexts from the hyper and quantized latent representations. The proposed architecture can be optimized end-to-end to fully exploit the context information and extract compact latent representation while reconstructing higher-quality images. Experimental results on four widely-used datasets showed that ConvNeXt-ChARM brings consistent and significant BD-rate (PSNR) reductions estimated on average to 5.24% and 1.22% over the versatile video coding (VVC) reference encoder (VTM-18.0) and the state-of-the-art learned image compression method SwinT-ChARM, respectively. Moreover, we provide model scaling studies to verify the computational efficiency of our approach and conduct several objective and subjective analyses to bring to the fore the performance gap between the next generation ConvNet, namely ConvNeXt, and Swin Transformer.
Motivated by the efficiency investigation of the Tranformer-based transform coding framework, namely SwinT-ChARM, we propose to enhance the latter, as first, with a more straightforward yet effective Tranformer-based channel-wise auto-regressive prior model, resulting in an absolute image compression transformer (ICT). Current methods that still rely on ConvNet-based entropy coding are limited in long-range modeling dependencies due to their local connectivity and an increasing number of architectural biases and priors. On the contrary, the proposed ICT can capture both global and local contexts from the latent representations and better parameterize the distribution of the quantized latents. Further, we leverage a learnable scaling module with a sandwich ConvNeXt-based pre/post-processor to accurately extract more compact latent representation while reconstructing higher-quality images. Extensive experimental results on benchmark datasets showed that the proposed adaptive image compression transformer (AICT) framework significantly improves the trade-off between coding efficiency and decoder complexity over the versatile video coding (VVC) reference encoder (VTM-18.0) and the neural codec SwinT-ChARM.
Recently, the performance of neural image compression (NIC) has steadily improved thanks to the last line of study, reaching or outperforming state-of-the-art conventional codecs. Despite significant progress, current NIC methods still rely on ConvNet-based entropy coding, limited in modeling long-range dependencies due to their local connectivity and the increasing number of architectural biases and priors, resulting in complex underperforming models with high decoding latency. Motivated by the efficiency investigation of the Tranformer-based transform coding framework, namely SwinT-ChARM, we propose to enhance the latter, as first, with a more straightforward yet effective Tranformer-based channel-wise auto-regressive prior model, resulting in an absolute image compression transformer (ICT). Through the proposed ICT, we can capture both global and local contexts from the latent representations and better parameterize the distribution of the quantized latents. Further, we leverage a learnable scaling module with a sandwich ConvNeXt-based pre-/post-processor to accurately extract more compact latent codes while reconstructing higher-quality images. Extensive experimental results on benchmark datasets showed that the proposed framework significantly improves the trade-off between coding efficiency and decoder complexity over the versatile video coding (VVC) reference encoder (VTM-18.0) and the neural codec SwinT-ChARM. Moreover, we provide model scaling studies to verify the computational efficiency of our approach and conduct several objective and subjective analyses to bring to the fore the performance gap between the adaptive image compression transformer (AICT) and the neural codec SwinT-ChARM.
Thanks to the remarkable advances in generative adversarial networks (GANs), it is becoming increasingly easy to generate/manipulate images. The existing works have mainly focused on deepfake in face images and videos. However, we are currently witnessing the emergence of fake satellite images, which can be misleading or even threatening to national security. Consequently, there is an urgent need to develop detection methods capable of distinguishing between real and fake satellite images. To advance the field, in this paper, we explore the suitability of several convolutional neural network (CNN) architectures for fake satellite image detection. Specifically, we benchmark four CNN models by conducting extensive experiments to evaluate their performance and robustness against various image distortions. This work allows the establishment of new baselines and may be useful for the development of CNN-based methods for fake satellite image detection.
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 paper, 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 (CNN) to extract discriminative features from image patches, which are then fed into two Recurrent Neural Networks (RNNs) for spatial and temporal pooling. Specifically, we use two Bi-directional Long Short Term Memory (Bi-LSTM) 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 video quality datasets show that 2BiVQA achieves high performance at a lower computational cost than state-of-the-art models. The source code of our 2BiVQA metric is made publicly available at: https://github.com/atelili/2BiVQA.
Changing the encoding parameters, in particular the video resolution, is a common practice before transcoding. To this end, streaming and broadcast platforms benefit from so-called bitrate ladders to determine the optimal resolution for given bitrates. However, the task of determining the bitrate ladder can usually be challenging as, on one hand, so-called fit-for-all static ladders would waste bandwidth, and on the other hand, fully specialized ladders are often not affordable in terms of computational complexity. In this paper, we propose an ML-based scheme for predicting the bitrate ladder based on the content of the video. The baseline of our solution predicts the bitrate ladder using two constituent methods, which require no encoding passes. To further enhance the performance of the constituent methods, we integrate a conditional ensemble method to aggregate their decisions, with a negligibly limited number of encoding passes. The experiment, carried out on the optimized software encoder implementation of the VVC standard, called VVenC, shows significant performance improvement. When compared to static bitrate ladder, the proposed method can offer about 13% bitrate reduction in terms of BD-BR with a negligible additional computational overhead. Conversely, when compared to the fully specialized bitrate ladder method, the proposed method can offer about 86% to 92% complexity reduction, at cost the of only 0.8% to 0.9% coding efficiency drop in terms of BD-BR.
The quality of patient care associated with diagnostic radiology is proportionate to a physician workload. Segmentation is a fundamental limiting precursor to diagnostic and therapeutic procedures. Advances in Machine Learning (ML) aim to increase diagnostic efficiency to replace single application with generalized algorithms. In Unsupervised Anomaly Detection (UAD), Convolutional Neural Network (CNN) based Autoencoders (AEs) and Variational Autoencoders (VAEs) are considered as a de facto approach for reconstruction based anomaly segmentation. Looking for anomalous regions in medical images is one of the main applications that use anomaly segmentation. The restricted receptive field in CNNs limit the CNN to model the global context and hence if the anomalous regions cover parts of the image, the CNN-based AEs are not capable to bring semantic understanding of the image. On the other hand, Vision Transformers (ViTs) have emerged as a competitive alternative to CNNs. It relies on the self-attention mechanism that is capable to relate image patches to each other. To reconstruct a coherent and more realistic image, in this work, we investigate Transformer capabilities in building AEs for reconstruction based UAD task. We focus on anomaly segmentation for Brain Magnetic Resonance Imaging (MRI) and present five Transformer-based models while enabling segmentation performance comparable or superior to State-of-The-Art (SOTA) models. The source code is available on Github https://github.com/ahmedgh970/Transformers_Unsupervised_Anomaly_Segmentation.git
In recent years, the global demand for high-resolution videos and the emergence of new multimedia applications have created the need for a new video coding standard. Hence, in July 2020 the Versatile Video Coding (VVC) standard was released providing up to 50% bit-rate saving for the same video quality compared to its predecessor High Efficiency Video Coding (HEVC). However, this bit-rate saving comes at the cost of a high computational complexity, particularly for live applications and on resource-constraint embedded devices. This paper presents two optimized VVC software decoders, named OpenVVC and Versatile Video deCoder (VVdeC), designed for low resources platforms. They exploit optimization techniques such as data level parallelism using Single Instruction Multiple Data (SIMD) instructions and functional level parallelism using frame, tile and slice-based parallelisms. Furthermore, a comparison in terms of decoding run time, energy and memory consumption between the two decoders is presented while targeting two different resource-constraint embedded devices. The results showed that both decoders achieve real-time decoding of Full High definition (FHD) resolution over the first platform using 8 cores and High-definition (HD) real-time decoding for the second platform using only 4 cores with comparable results in terms of average consumed energy: around 26 J and 15 J for the 8 cores and 4 cores embedded platforms, respectively. Regarding the memory usage, OpenVVC showed better results with less average maximum memory consumed during run time compared to VVdeC.