Real-Time Super-Resolution for Real-World Images on Mobile Devices

Image Super-Resolution (ISR), which aims at recovering High-Resolution (HR) images from the corresponding Low-Resolution (LR) counterparts. Although recent progress in ISR has been remarkable. However, they are way too computationally intensive to be deployed on edge devices, since most of the recent approaches are deep learning-based. Besides, these methods always fail in real-world scenes, since most of them adopt a simple fixed "ideal" bicubic downsampling kernel from high-quality images to construct LR/HR training pairs which may lose track of frequency-related details. In this work, an approach for real-time ISR on mobile devices is presented, which is able to deal with a wide range of degradations in real-world scenarios. Extensive experiments on traditional super-resolution datasets (Set5, Set14, BSD100, Urban100, Manga109, DIV2K) and real-world images with a variety of degradations demonstrate that our method outperforms the state-of-art methods, resulting in higher PSNR and SSIM, lower noise and better visual quality. Most importantly, our method achieves real-time performance on mobile or edge devices.

MM-ViT: Multi-Modal Video Transformer for Compressed Video Action Recognition

This paper presents a pure transformer-based approach, dubbed the Multi-Modal Video Transformer (MM-ViT), for video action recognition. Different from other schemes which solely utilize the decoded RGB frames, MM-ViT operates exclusively in the compressed video domain and exploits all readily available modalities, i.e., I-frames, motion vectors, residuals and audio waveform. In order to handle the large number of spatiotemporal tokens extracted from multiple modalities, we develop several scalable model variants which factorize self-attention across the space, time and modality dimensions. In addition, to further explore the rich inter-modal interactions and their effects, we develop and compare three distinct cross-modal attention mechanisms that can be seamlessly integrated into the transformer building block. Extensive experiments on three public action recognition benchmarks (UCF-101, Something-Something-v2, Kinetics-600) demonstrate that MM-ViT outperforms the state-of-the-art video transformers in both efficiency and accuracy, and performs better or equally well to the state-of-the-art CNN counterparts with computationally-heavy optical flow.

RSCA: Real-time Segmentation-based Context-Aware Scene Text Detection

Segmentation-based scene text detection methods have been widely adopted for arbitrary-shaped text detection recently, since they make accurate pixel-level predictions on curved text instances and can facilitate real-time inference without time-consuming processing on anchors. However, current segmentation-based models are unable to learn the shapes of curved texts and often require complex label assignments or repeated feature aggregations for more accurate detection. In this paper, we propose RSCA: a Real-time Segmentation-based Context-Aware model for arbitrary-shaped scene text detection, which sets a strong baseline for scene text detection with two simple yet effective strategies: Local Context-Aware Upsampling and Dynamic Text-Spine Labeling, which model local spatial transformation and simplify label assignments separately. Based on these strategies, RSCA achieves state-of-the-art performance in both speed and accuracy, without complex label assignments or repeated feature aggregations. We conduct extensive experiments on multiple benchmarks to validate the effectiveness of our method. RSCA-640 reaches 83.9% F-measure at 48.3 FPS on CTW1500 dataset.

Real-Time Video Super-Resolution on Smartphones with Deep Learning, Mobile AI 2021 Challenge: Report

Video super-resolution has recently become one of the most important mobile-related problems due to the rise of video communication and streaming services. While many solutions have been proposed for this task, the majority of them are too computationally expensive to run on portable devices with limited hardware resources. To address this problem, we introduce the first Mobile AI challenge, where the target is to develop an end-to-end deep learning-based video super-resolution solutions that can achieve a real-time performance on mobile GPUs. The participants were provided with the REDS dataset and trained their models to do an efficient 4X video upscaling. The runtime of all models was evaluated on the OPPO Find X2 smartphone with the Snapdragon 865 SoC capable of accelerating floating-point networks on its Adreno GPU. The proposed solutions are fully compatible with any mobile GPU and can upscale videos to HD resolution at up to 80 FPS while demonstrating high fidelity results. A detailed description of all models developed in the challenge is provided in this paper.

AIM 2020 Challenge on Efficient Super-Resolution: Methods and Results

This paper reviews the AIM 2020 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The challenge task was to super-resolve an input image with a magnification factor x4 based on a set of prior examples of low and corresponding high resolution images. The goal is to devise a network that reduces one or several aspects such as runtime, parameter count, FLOPs, activations, and memory consumption while at least maintaining PSNR of MSRResNet. The track had 150 registered participants, and 25 teams submitted the final results. They gauge the state-of-the-art in efficient single image super-resolution.

GIA-Net: Global Information Aware Network for Low-light Imaging

It is extremely challenging to acquire perceptually plausible images under low-light conditions due to low SNR. Most recently, U-Nets have shown promising results for low-light imaging. However, vanilla U-Nets generate images with artifacts such as color inconsistency due to the lack of global color information. In this paper, we propose a global information aware (GIA) module, which is capable of extracting and integrating the global information into the network to improve the performance of low-light imaging. The GIA module can be inserted into a vanilla U-Net with negligible extra learnable parameters or computational cost. Moreover, a GIA-Net is constructed, trained and evaluated on a large scale real-world low-light imaging dataset. Experimental results show that the proposed GIA-Net outperforms the state-of-the-art methods in terms of four metrics, including deep metrics that measure perceptual similarities. Extensive ablation studies have been conducted to verify the effectiveness of the proposed GIA-Net for low-light imaging by utilizing global information.

Residual Frames with Efficient Pseudo-3D CNN for Human Action Recognition

Human action recognition is regarded as a key cornerstone in domains such as surveillance or video understanding. Despite recent progress in the development of end-to-end solutions for video-based action recognition, achieving state-of-the-art performance still requires using auxiliary hand-crafted motion representations, e.g., optical flow, which are usually computationally demanding. In this work, we propose to use residual frames (i.e., differences between adjacent RGB frames) as an alternative "lightweight" motion representation, which carries salient motion information and is computationally efficient. In addition, we develop a new pseudo-3D convolution module which decouples 3D convolution into 2D and 1D convolution. The proposed module exploits residual information in the feature space to better structure motions, and is equipped with a self-attention mechanism that assists to recalibrate the appearance and motion features. Empirical results confirm the efficiency and effectiveness of residual frames as well as the proposed pseudo-3D convolution module.

NTIRE 2020 Challenge on Real-World Image Super-Resolution: Methods and Results

This paper reviews the NTIRE 2020 challenge on real world super-resolution. It focuses on the participating methods and final results. The challenge addresses the real world setting, where paired true high and low-resolution images are unavailable. For training, only one set of source input images is therefore provided along with a set of unpaired high-quality target images. In Track 1: Image Processing artifacts, the aim is to super-resolve images with synthetically generated image processing artifacts. This allows for quantitative benchmarking of the approaches \wrt a ground-truth image. In Track 2: Smartphone Images, real low-quality smart phone images have to be super-resolved. In both tracks, the ultimate goal is to achieve the best perceptual quality, evaluated using a human study. This is the second challenge on the subject, following AIM 2019, targeting to advance the state-of-the-art in super-resolution. To measure the performance we use the benchmark protocol from AIM 2019. In total 22 teams competed in the final testing phase, demonstrating new and innovative solutions to the problem.

Residual Channel Attention Generative Adversarial Network for Image Super-Resolution and Noise Reduction

Image super-resolution is one of the important computer vision techniques aiming to reconstruct high-resolution images from corresponding low-resolution ones. Most recently, deep learning-based approaches have been demonstrated for image super-resolution. However, as the deep networks go deeper, they become more difficult to train and more difficult to restore the finer texture details, especially under real-world settings. In this paper, we propose a Residual Channel Attention-Generative Adversarial Network(RCA-GAN) to solve these problems. Specifically, a novel residual channel attention block is proposed to form RCA-GAN, which consists of a set of residual blocks with shortcut connections, and a channel attention mechanism to model the interdependence and interaction of the feature representations among different channels. Besides, a generative adversarial network (GAN) is employed to further produce realistic and highly detailed results. Benefiting from these improvements, the proposed RCA-GAN yields consistently better visual quality with more detailed and natural textures than baseline models; and achieves comparable or better performance compared with the state-of-the-art methods for real-world image super-resolution.

Gradient-Coherent Strong Regularization for Deep Neural Networks

Deep neural networks are often prone to over-fitting with their numerous parameters, so regularization plays an important role in generalization. L1 and L2 regularizers are common regularization tools in machine learning with their simplicity and effectiveness. However, we observe that imposing strong L1 or L2 regularization on deep neural networks with stochastic gradient descent easily fails, which limits the generalization ability of the underlying neural networks. To understand this phenomenon, we first investigate how and why learning fails when strong regularization is imposed on deep neural networks. We then propose a novel method, gradient-coherent strong regularization, which imposes regularization only when the gradients are kept coherent in the presence of strong regularization. Experiments are performed with multiple deep architectures on three benchmark data sets for image recognition. Experimental results show that our proposed approach indeed endures strong regularization and significantly improves both accuracy and compression, which could not be achieved otherwise.