Image datasets with paired mean opinion scores (MOS) enable the quantization of the perceptual differences between images and are of great significance for images taken underwater. In our previous research, a pairwise label underwater image quality subjective ranking (PLUIQR) method was proposed. In this paper, we take a further step by designing a post-reliability verification for the PLUIQR and releasing a publicly accessible underwater image quality dataset called PCUID. For the raw paired voting, steps including a triangular cycle error (TCE) criterion and reprocessing of the dispute data are performed. Meanwhile, the group maximum differentiation (gMAD) method is used to evaluate the performance of underwater quality assessment methods (UIQA) on images with similar quality. That illustrates the proposed dataset enabling UIQA methods to their judgment toward subtle quality differences. The dataset is available at https://github.com/JOU-UIP/PCUID.

ABSTRACT Underwater image quality evaluation (UIQE) is crucial in improving image processing techniques and optimizing the design of the imaging system to obtain object information more accurately. However, existing UIQE methods are designed based on limited images or consider only a few natural scene statistics (NSS) metrics, lacking consideration for generalization across various underwater imaging applications. In this paper, an in‐depth review of the existing UIQE methods based on evaluation operations is provided, emphasizing the bias present when evaluating UIQE methods using individual metrics. To address this, a novel metric called quadrilateral datum evaluation (QDE) is designed for UIQE methods. It comprehensively considers robustness across different datasets, as well as correlation and ranking consistency with mean opinion scores (MOS). This is the first solution to measure an UIQE method from an all‐encompassing visual perspective. By using QDE, UIQE methods characterized by greater feature strength and small imbalance demonstrate good consistency and robustness across multiple aspects, providing a basis for the design of UIQE methods.

With the development of computer vision technology, unsupervised depth estimation from single images has experienced significant advancements under normal weather conditions, demonstrating highly promising results. Nevertheless, its efficacy in estimating depth under less-than-optimal weather conditions, particularly those characterized by fog, continues to pose substantial challenges. In this paper, we propose FoggyDepth that is designed to utilize channel-wise Fourier transform to remedy this limitation. Specifically, to relieve the problem of photometric consistency assumption not holding in foggy scenes within the unsupervised framework, we employ a channel-dimension Fourier transform to obtain channel global statistical information, thereby enhancing the discriminative ability of global representation. Meanwhile, we generate a series of foggy scene samples corresponding to normal training samples and use them for self-supervised training to guide the model to accurately recover depth in foggy conditions. In addition, to further improve the model performance, we utilize a non-local network to capture long-range spatial dependencies in depth estimation. Comprehensive evaluations conducted on the Oxford RobotCar, nuScenes, and Driving Stereo datasets substantiate the precision and reliability of our proposed method. Through a meticulous comparison with existing leading-edge algorithms in depth estimation, our approach demonstrates superior performance, both qualitatively and quantitatively.

In the pursuit of robust and generalizable environment perception and language understanding, the ubiquitous challenge of dataset bias continues to plague vision-and-language navigation (VLN) agents, hindering their performance in unseen environments. This paper introduces the generalized cross-modal causal transformer (GOAT), a pioneering solution rooted in the paradigm of causal inference. By delving into both observable and unobservable confounders within vision, language, and history, we propose the back-door and front-door adjustment causal learning (BACL and FACL) modules to promote unbiased learning by comprehensively mitigating potential spurious correlations. Additionally, to capture global confounder features, we propose a cross-modal feature pooling (CFP) module supervised by contrastive learning, which is also shown to be effective in improving cross-modal representations during pretraining. Extensive experiments across multiple VLN datasets (R2R, REVERIE, RxR, and SOON) underscore the superiority of our proposed method over previous state-of-the-art approaches. Code is available at https://github.com/CrystalSixone/VLN-GOAT.

Monocular Bird’s Eye View (BEV) semantic segmentation is critical for autonomous driving for its inherent advantages in spatial representation and downstream tasks. However, it is challenging to simultaneously learn view transformation and pixel-wise classification. Previous works suffer from non-flat region distortion, distant depth ambiguity, and visual occlusion. To address these aforementioned concerns, we propose dual-branch view transformation (DVT), a novel framework for monocular BEV semantic segmentation. Our method consists of: (i) A dual-branch view transformation to decouple features into flat region and non-flat region and process them independently. (ii) A depth-aware weighting method to make the model pay more attention to the distant depth. (iii) An auxiliary task to introduce more inductive biases to alleviate the inaccuracy caused by visual occlusion. Furthermore, we design a class-aware weighting method to address the class and size imbalance of datasets. Experimental results on nuScenes and KITTI-360 datasets demonstrate that DVT outperforms previous state-of-the-art (SOTA). Our codes are available at https://github.com/MrPicklesGG/DVT.

The varying environmental conditions pose challenges to existing object detection methods as they lead to changes in the overall feature distribution of images. Underwater images are particularly susceptible to environmental conditions changes, resulting in phenomena like color deviation. This paper propose an object detection model, FIOD-VUE, which focuses on invariant information across different underwater environments to enhance the model’s generalization capability. Inspired by frequency domain analysing, we design a Frequency-Invariant Attention (FIA) module. This module use frequency filters to focus on specific frequency signals, i.e., cross-domain invariant information. Additionally, we design the Multi-scale Image-level Feature Alignment (MIFA) to adaptively adjust the frequency filters in the FIA and assist the backbone in extracting domain-confusion features. Through adversarial training, the distribution gap between the source domain and target domain is reduced. To enrich the domain shift database, we also afford an HD-Deepfish dataset. Numerous experiments on the S-UODAC2020 and the HD-Deepfish datasets were executed and yielded impressive results, with average precision (AP) scores of 56.8% and 37.1%, respectively, surpassing the performance of the existing underwater object detection (UOD) models. The link of the code is released at:https://github.com/JOU-UIP/FIOD-VUE.