CPDM: Content-Preserving Diffusion Model for Underwater Image Enhancement

Underwater image enhancement (UIE) is challenging since image degradation in aquatic environments is complicated and changing over time. Existing mainstream methods rely on either physical-model or data-driven, suffering from performance bottlenecks due to changes in imaging conditions or training instability. In this article, we make the first attempt to adapt the diffusion model to the UIE task and propose a Content-Preserving Diffusion Model (CPDM) to address the above challenges. CPDM first leverages a diffusion model as its fundamental model for stable training and then designs a content-preserving framework to deal with changes in imaging conditions. Specifically, we construct a conditional input module by adopting both the raw image and the difference between the raw and noisy images as the input, which can enhance the model's adaptability by considering the changes involving the raw images in underwater environments. To preserve the essential content of the raw images, we construct a content compensation module for content-aware training by extracting low-level features from the raw images. Extensive experimental results validate the effectiveness of our CPDM, surpassing the state-of-the-art methods in terms of both subjective and objective metrics.

RFI-DRUnet: Restoring dynamic spectra corrupted by radio frequency interference -- Application to pulsar observations

Radio frequency interference (RFI) have been an enduring concern in radio astronomy, particularly for the observations of pulsars which require high timing precision and data sensitivity. In most works of the literature, RFI mitigation has been formulated as a detection task that consists of localizing possible RFI in dynamic spectra. This strategy inevitably leads to a potential loss of information since parts of the signal identified as possibly RFI-corrupted are generally not considered in the subsequent data processing pipeline. Conversely, this work proposes to tackle RFI mitigation as a joint detection and restoration that allows parts of the dynamic spectrum affected by RFI to be not only identified but also recovered. The proposed supervised method relies on a deep convolutional network whose architecture inherits the performance reached by a recent yet popular image-denoising network. To train this network, a whole simulation framework is built to generate large data sets according to physics-inspired and statistical models of the pulsar signals and of the RFI. The relevance of the proposed approach is quantitatively assessed by conducting extensive experiments. In particular, the results show that the restored dynamic spectra are sufficiently reliable to estimate pulsar times-of-arrivals with an accuracy close to the one that would be obtained from RFI-free signals.

Hybrid Video Diffusion Models with 2D Triplane and 3D Wavelet Representation

Generating high-quality videos that synthesize desired realistic content is a challenging task due to their intricate high-dimensionality and complexity of videos. Several recent diffusion-based methods have shown comparable performance by compressing videos to a lower-dimensional latent space, using traditional video autoencoder architecture. However, such method that employ standard frame-wise 2D and 3D convolution fail to fully exploit the spatio-temporal nature of videos. To address this issue, we propose a novel hybrid video diffusion model, called HVDM, which can capture spatio-temporal dependencies more effectively. The HVDM is trained by a hybrid video autoencoder which extracts a disentangled representation of the video including: (i) a global context information captured by a 2D projected latent (ii) a local volume information captured by 3D convolutions with wavelet decomposition (iii) a frequency information for improving the video reconstruction. Based on this disentangled representation, our hybrid autoencoder provide a more comprehensive video latent enriching the generated videos with fine structures and details. Experiments on video generation benchamarks (UCF101, SkyTimelapse, and TaiChi) demonstrate that the proposed approach achieves state-of-the-art video generation quality, showing a wide range of video applications (e.g., long video generation, image-to-video, and video dynamics control).

Improving Language Understanding from Screenshots

An emerging family of language models (LMs), capable of processing both text and images within a single visual view, has the promise to unlock complex tasks such as chart understanding and UI navigation. We refer to these models as screenshot language models. Despite their appeal, existing screenshot LMs substantially lag behind text-only models on language understanding tasks. To close this gap, we adopt a simplified setting where the model inputs are plain-text-rendered screenshots, and we focus on improving the text ability of screenshot LMs. We propose a novel Patch-and-Text Prediction (PTP) objective, which masks and recovers both image patches of screenshots and text within screenshots. We also conduct extensive ablation studies on masking rates and patch sizes, as well as designs for improving training stability. Our pre-trained model, while solely taking visual inputs, achieves comparable performance with BERT on 6 out of 8 GLUE tasks (within 2%) and improves up to 8% over prior work. Additionally, we extend PTP to train autoregressive screenshot LMs and demonstrate its effectiveness--our models can significantly reduce perplexity by utilizing the screenshot context. Together, we hope our findings can inspire future research on developing powerful screenshot LMs and extending their reach to broader applications.

Deep Spectral Improvement for Unsupervised Image Instance Segmentation

Deep spectral methods reframe the image decomposition process as a graph partitioning task by extracting features using self-supervised learning and utilizing the Laplacian of the affinity matrix to obtain eigensegments. However, instance segmentation has received less attention compared to other tasks within the context of deep spectral methods. This paper addresses the fact that not all channels of the feature map extracted from a self-supervised backbone contain sufficient information for instance segmentation purposes. In fact, Some channels are noisy and hinder the accuracy of the task. To overcome this issue, this paper proposes two channel reduction modules: Noise Channel Reduction (NCR) and Deviation-based Channel Reduction (DCR). The NCR retains channels with lower entropy, as they are less likely to be noisy, while DCR prunes channels with low standard deviation, as they lack sufficient information for effective instance segmentation. Furthermore, the paper demonstrates that the dot product, commonly used in deep spectral methods, is not suitable for instance segmentation due to its sensitivity to feature map values, potentially leading to incorrect instance segments. A new similarity metric called Bray-Curtis over Chebyshev (BoC) is proposed to address this issue. It takes into account the distribution of features in addition to their values, providing a more robust similarity measure for instance segmentation. Quantitative and qualitative results on the Youtube-VIS2019 dataset highlight the improvements achieved by the proposed channel reduction methods and the use of BoC instead of the conventional dot product for creating the affinity matrix. These improvements are observed in terms of mean Intersection over Union and extracted instance segments, demonstrating enhanced instance segmentation performance. The code is available on: https://github.com/farnooshar/SpecUnIIS

FD-Vision Mamba for Endoscopic Exposure Correction

In endoscopic imaging, the recorded images are prone to exposure abnormalities, so maintaining high-quality images is important to assist healthcare professionals in performing decision-making. To overcome this issue, We design a frequency-domain based network, called FD-Vision Mamba (FDVM-Net), which achieves high-quality image exposure correction by reconstructing the frequency domain of endoscopic images. Specifically, inspired by the State Space Sequence Models (SSMs), we develop a C-SSM block that integrates the local feature extraction ability of the convolutional layer with the ability of the SSM to capture long-range dependencies. A two-path network is built using C-SSM as the basic function cell, and these two paths deal with the phase and amplitude information of the image, respectively. Finally, a degraded endoscopic image is reconstructed by FDVM-Net to obtain a high-quality clear image. Extensive experimental results demonstrate that our method achieves state-of-the-art results in terms of speed and accuracy, and it is noteworthy that our method can enhance endoscopic images of arbitrary resolution. The URL of the code is \url{https://github.com/zzr-idam/FDVM-Net}.

DiCoM -- Diverse Concept Modeling towards Enhancing Generalizability in Chest X-Ray Studies

Chest X-Ray (CXR) is a widely used clinical imaging modality and has a pivotal role in the diagnosis and prognosis of various lung and heart related conditions. Conventional automated clinical diagnostic tool design strategies relying on radiology reads and supervised learning, entail the cumbersome requirement of high quality annotated training data. To address this challenge, self-supervised pre-training has proven to outperform supervised pre-training in numerous downstream vision tasks, representing a significant breakthrough in the field. However, medical imaging pre-training significantly differs from pre-training with natural images (e.g., ImageNet) due to unique attributes of clinical images. In this context, we introduce Diverse Concept Modeling (DiCoM), a novel self-supervised training paradigm that leverages a student teacher framework for learning diverse concepts and hence effective representation of the CXR data. Hence, expanding beyond merely modeling a single primary label within an image, instead, effectively harnessing the information from all the concepts inherent in the CXR. The pre-trained model is subsequently fine-tuned to address diverse domain-specific tasks. Our proposed paradigm consistently demonstrates robust performance across multiple downstream tasks on multiple datasets, highlighting the success and generalizability of the pre-training strategy. To establish the efficacy of our methods we analyze both the power of learned representations and the speed of convergence (SoC) of our models. For diverse data and tasks, DiCoM is able to achieve in most cases better results compared to other state-of-the-art pre-training strategies. This when combined with the higher SoC and generalization capabilities positions DiCoM to be established as a foundation model for CXRs, a widely used imaging modality.

HistoSegCap: Capsules for Weakly-Supervised Semantic Segmentation of Histological Tissue Type in Whole Slide Images

Digital pathology involves converting physical tissue slides into high-resolution Whole Slide Images (WSIs), which pathologists analyze for disease-affected tissues. However, large histology slides with numerous microscopic fields pose challenges for visual search. To aid pathologists, Computer Aided Diagnosis (CAD) systems offer visual assistance in efficiently examining WSIs and identifying diagnostically relevant regions. This paper presents a novel histopathological image analysis method employing Weakly Supervised Semantic Segmentation (WSSS) based on Capsule Networks, the first such application. The proposed model is evaluated using the Atlas of Digital Pathology (ADP) dataset and its performance is compared with other histopathological semantic segmentation methodologies. The findings underscore the potential of Capsule Networks in enhancing the precision and efficiency of histopathological image analysis. Experimental results show that the proposed model outperforms traditional methods in terms of accuracy and the mean Intersection-over-Union (mIoU) metric.

Multilingual Text-to-Image Generation Magnifies Gender Stereotypes and Prompt Engineering May Not Help You

Text-to-image generation models have recently achieved astonishing results in image quality, flexibility, and text alignment and are consequently employed in a fast-growing number of applications. Through improvements in multilingual abilities, a larger community now has access to this kind of technology. Yet, as we will show, multilingual models suffer similarly from (gender) biases as monolingual models. Furthermore, the natural expectation is that these models will provide similar results across languages, but this is not the case and there are important differences between languages. Thus, we propose a novel benchmark MAGBIG intending to foster research in multilingual models without gender bias. We investigate whether multilingual T2I models magnify gender bias with MAGBIG. To this end, we use multilingual prompts requesting portrait images of persons of a certain occupation or trait (using adjectives). Our results show not only that models deviate from the normative assumption that each gender should be equally likely to be generated, but that there are also big differences across languages. Furthermore, we investigate prompt engineering strategies, i.e. the use of indirect, neutral formulations, as a possible remedy for these biases. Unfortunately, they help only to a limited extent and result in worse text-to-image alignment. Consequently, this work calls for more research into diverse representations across languages in image generators.

Self-Play Fine-Tuning of Diffusion Models for Text-to-Image Generation

Fine-tuning Diffusion Models remains an underexplored frontier in generative artificial intelligence (GenAI), especially when compared with the remarkable progress made in fine-tuning Large Language Models (LLMs). While cutting-edge diffusion models such as Stable Diffusion (SD) and SDXL rely on supervised fine-tuning, their performance inevitably plateaus after seeing a certain volume of data. Recently, reinforcement learning (RL) has been employed to fine-tune diffusion models with human preference data, but it requires at least two images ("winner" and "loser" images) for each text prompt. In this paper, we introduce an innovative technique called self-play fine-tuning for diffusion models (SPIN-Diffusion), where the diffusion model engages in competition with its earlier versions, facilitating an iterative self-improvement process. Our approach offers an alternative to conventional supervised fine-tuning and RL strategies, significantly improving both model performance and alignment. Our experiments on the Pick-a-Pic dataset reveal that SPIN-Diffusion outperforms the existing supervised fine-tuning method in aspects of human preference alignment and visual appeal right from its first iteration. By the second iteration, it exceeds the performance of RLHF-based methods across all metrics, achieving these results with less data.