Transmitting images through narrow-bandwidth underwater acoustic channels presents a significant challenge. However, for many applications, transmitting the entire image is not necessary. Instead, conveying only the critical information is typically sufficient to ensure task success. To address this, we have introduced a preprocessing strategy at the image acquisition stage, which involves down-sampling or feature extraction to distill the data into its essential components. This strategy is guided by utility considerations to ensure that the transmitted data is both concise and informative. However, existing quality evaluation methods are inadequate for providing accurate utility assessments and fail to simultaneously evaluate the utility of both images and features within a unified framework. To bridge this gap in underwater object detection (one of the typical underwater task), we have developed an underwater image utility dataset and introduced the Lightweight and Unified Utility for Underwater Image Evaluation (LU3IE). LU3IE leverages features derived from multiple detection models to enhance generalization. It then refines these features through dimensionality reduction and commonality extraction, and simplifies complexity through the use of knowledge distillation. Experimental results validate the effectiveness and scalability of LU3IE in accurately assessing the utility of underwater images.

Most current Handwritten Mathematical Expression Recognition (HMER) methods employ an attention-based encoder-decoder framework, which generates LaTeX sequences from the given images, following the paradigm of predicting "one-by-one". However, this paradigm may have some challenges: 1) without considering the connectivity between characters, the prior information in the prediction process will be ignored inadvertently, especially implicit information, such as " " and " ˆ ". 2) Some characters of high similarities, such as "6/b" and "o/O", will have negative effects on prediction results. To solve these issues, we propose a simple but effective Character Relationship Refinement Network (CRRN), which consists of Joint Character Learning (JCL) and Character Refinement Mask (CRM). Specifically, JCL calculates the relationship probability between characters and uses them to improve prediction accuracy. CRM takes the character confidence coefficient in a coarse-to-fine way that can reassign the weights of all characters to improve model discriminability on easily confused characters. With the collaboration of both modules, our proposed CRRN can outperform the state-of-the-art on popular datasets.

The limited bandwidth of underwater acoustic channels poses a challenge to the efficiency of multimedia information transmission. To improve efficiency, the system aims to transmit less data while maintaining image utility at the receiving end. Although assessing utility within compressed information is essential, the current methods exhibit limitations in addressing utility-driven quality assessment. Therefore, this paper introduces a Distillation-based Compacted Information Quality assessment metric (DCIQ) for utility-oriented quality evaluation in the context of underwater machine vision. This method is conducted in the Utility-oriented compacted Image Quality Dataset (UIQD) that contains utility qualities of reference images and their corresponding compressed information at different levels. The utility score is derived from the average confidence of various object detection models. In DCIQ, utility features of compacted information are acquired through transfer learning and mapped using a Transformer. Besides, we propose a utility-oriented cross-model feature fusion mechanism to address different detection algorithm preferences. After that, a utility-oriented feature quality measure assesses compacted feature utility. Finally, we utilize distillation to compress the model by reducing its parameters by 55%. Experiment results effectively demonstrate that our proposed DCIQ can predict utility-oriented quality within compressed underwater information

Recently, many compression algorithms are applied to decrease the cost of video storage and transmission. This will introduce undesirable artifacts, which severely degrade visual quality. Therefore, Video Compression Artifacts Removal (VCAR) aims at reconstructing a high-quality video from its corrupted version of compression. Generally, this task is considered as a vision-related instead of media-related problem. In vision-related research, the visual quality has been significantly improved while the computational complexity and bitrate issues are less considered. In this work, we review the performance constraints of video coding and transfer to evaluate the VCAR outputs. Based on the analyses, we propose a Spatial-Temporal Attention-Guided Enhancement Network (STAGE-Net). First, we employ dynamic filter processing, instead of conventional optical flow method, to reduce the computational cost of VCAR. Second, we introduce self-attention mechanism to design Sequential Residual Attention Blocks (SRABs) to improve visual quality of enhanced video frames with bitrate constraints. Both quantitative and qualitative experimental results have demonstrated the superiority of our proposed method, which achieves high visual qualities and low computational costs.

Low-light imaging task aims to approximate low-light scenes as perceived by human eyes. Existing methods usually pursue higher brightness, resulting in unrealistic exposure. Inspired by Human Vision System (HVS), where rods perceive more lights while cones perceive more colors, we propose a Low-light Degradation Rectify Model (LDRM) with color-monochrome cameras to solve this problem. First, we propose to use a low-ISO color camera and a high-ISO monochrome camera for low-light imaging under short-exposure of less than 0.1s. Short-exposure could avoid motion blurriness, while monochrome camera captures more photons than color camera. By mimicing HVS, this capture system could benefit low-light imaging. Second, we propose an LDRM model to fuse the color-monochrome image pair into a high-quality image. In this model, we separately restore UV and Y channels through chrominance and luminance branches and use monochrome image to guide the restoration of luminance. We also propose a latent code embedding method to improve the restorations of both branches. Third, we create a Low-light Color-Monochrome benchmark (LCM), including both synthetic and real-world datasets, to examine low-light imaging quality of LDRM and the state-of-the-art methods. Experimental results demonstrate the superior performance of LDRM with visually pleasing results. Codes and datasets are available at https://github.com/StephenLinn/LDRM.

Single-image desnowing aims at depressing snowflake noises while preserving a clean background. Existing methods usually mask the locations of noises and remove them in RGB color space. In this paper, we rethink this problem by investigating the impacts of color space selection. Theoretical analysis and experiments reveal that the feature of snowflake noises exhibit different distributions in different color spaces. In particular, these noises are barely seen in Hue channel, which inspires us to recover global structure and texture information of the clean background from Hue channel. More low-frequency information is also found in the Hue channel. With these observations, we propose a novel Hybrid-Color-Space-based Desnowing Network (HCSD-Net). The proposed HCSD-Net extracts low-frequency and high-frequency features in Hue channel and RGB color space, respectively. After that, it utilizes a multi-scale fusion module to enhance high-frequency details at a small feature resolution. These details are further used to supervise and supplement the background information. Extensive experiments demonstrate that our proposed HCSD-Net outperforms state-of-the-art methods on various synthetic and real-world desnowing datasets. Codes are available at https://github.com/ttz-rainbow/HCSD-Net.

Deep learning technologies have shown their advantages in Single Image Rain Removal (SIRR) tasks. However, the derained results of most methods are limited to some challenges. First, due to the lack of real-world rainy/clean image pairs, many methods seriously rely on the labeled synthetic training images and will not effectively remove complex rain streaks in real-world scenarios. Second, most existing SIRR models require high computing power, which considerably limits their real-world applications. To address these issues, we propose a Lightweight Semi-supervised Network (LSNet) for SIRR. Our LSNet utilizes a compact semi-supervised framework to improve generalization ability in real-world rainy images removal. Meanwhile, in our semi-supervised framework, we also design a cascaded sub-network, which progressively removes complex rain streaks via a multi-stage manner. Specially, the multi-stage manner is based on a series of cascaded blocks, where we conduct recursive learning strategy to reduce model parameters. Extensive experimental results demonstrate that our method achieves comparable performance to the state-of-the-arts while has fewer parameters.

Images collected in low-light environments usually suffer from multiple, non-uniform distributed distortions, including local dark, dim light, backlit and so on. In this paper, we propose a Stage-Transformer-Guided Network (STGNet) that effectively handles region-specific distributions and enhance diverse low-light images. Specifically, our STGNet adopts a multi-stage way to progressively learn hierarchical features that benefit the robustness of our model. At each stage, we design an efficient transformer with horizontal and vertical attentions that jointly capture degradation distributions with different magnitudes and orientations. We also introduce learnable degradation queries to adaptively select task-specific features of degradations for enhancement. In addition, we design a histogram loss for enhancement and combine it with other loss functions, in order to exploit both global contrast and local details during network training. Benefiting from the above contributions, our STGNet achieves the state-of-the-art performances on both synthetic and real-world datasets.

Snow removal aims to locate snow areas and recover clean images without repairing traces. Unlike the regularity and semitransparency of rain, snow with various patterns and degradations seriously occludes the background. As a result, the state-of-the-art snow removal methods usually retains a large parameter size. In this paper, we propose a lightweight but high-efficient snow removal network called Laplace Mask Query Transformer (LMQFormer). Firstly, we present a Laplace-VQVAE to generate a coarse mask as prior knowledge of snow. Instead of using the mask in dataset, we aim at reducing both the information entropy of snow and the computational cost of recovery. Secondly, we design a Mask Query Transformer (MQFormer) to remove snow with the coarse mask, where we use two parallel encoders and a hybrid decoder to learn extensive snow features under lightweight requirements. Thirdly, we develop a Duplicated Mask Query Attention (DMQA) that converts the coarse mask into a specific number of queries, which constraint the attention areas of MQFormer with reduced parameters. Experimental results in popular datasets have demonstrated the efficiency of our proposed model, which achieves the state-of-the-art snow removal quality with significantly reduced parameters and the lowest running time. Codes and models are available athttps://github.com/StephenLinn/LMQFormer.

Rain streaks usually result in severe image visual degradation and foreground occlusion, affecting the quality of computer tasks in outdoor scenes. Currently, the mainstream methods in single-image deraining are based on data-driven. However, the deep learning network could be imperfect, with limited power for learning the global information from rain streaks all over the map. In order to solve this problem, we proposed a novel Hierarchical Distillation Network (HD-Net). In this network, Hierarchical Feature Extraction Block (HFEB) can fully utilize the Transformer's learning ability in high-level features, integrate local detail extraction and global structure representation, and compensate for the weakness of the Convolutional Neural Network (CNN), which is overattentive to the underlying image features. Furthermore, the Distillation-Calibration Block (DCB) are adopted to avoid feature redundancy during model training and calibrate the channel and spatial information through the feature transmission, which could significantly improve the learning efficiency. Finally, the experiment results show that our model performs better than traditional CNN models and state-of-the-art methods.

Single image desnowing is an important and challenge task for lots of computer vision applications, such as visual tracking and video surveillance. Although existing deep learning-based methods have achieved promising results, most of them rely on the local deep features and neglect global relationship information between the local regions. Therefore, inevitably leading to over-smooth or detail loss results. To solve this issue, we design a UNet-based end-to-end architecture for image desnowing. Specially, to better characterize global information and preserve image detail, we combine Window-based Self-Attention (WSA) transformer block with Residue Spatial Attention (RSA) to build basic unit of our network. Besides, to protect the structure of the image effectively, we also introduce a Residue Channel (RC) loss to guide high-quality image restoration. Extensive experimental results on both synthetic and real-world datasets demonstrate that the proposed model achieves new state-of-the-art results.

Due to light scatter and absorption in waterbody, underwater imaging can be easily impaired with low contrast and visual distortion. The resulting images are often unable to meet the quality requirements of human perception and computer processing. Therefore, Underwater Image Enhancement (UIE) has been attracting extensive research efforts. Although deep learning has demonstrated its great success in many vision tasks, its huge amounts of parameters and computations are not conducive to UIE in resource-limited scenarios. In this paper, we address this issue by proposing a Lightweight Cascaded Network (LCNet) based on Laplacian image pyramids. At each pyramid level, we implement cascaded blocks upon a residual network. Specifically, high quality residuals can be progressively predicted with significantly reduced complexity in a coarse-to-fine fashion. Furthermore, these sub-networks are recursively nested to build our LCNet, thereby reducing the overall computational complexity with reused parameters. Extensive experiments demonstrate that the proposed method performs favorably against the state-of-the-arts in terms of visual quality, model parameters and complexity.