Not a statistical accident: framing bias as a semiotic property of image datasets

The informed sampler: A discriminative approach to Bayesian inference in generative computer vision models

Computer vision systems are increasingly expanding their application in visual data analysis. Model training methods are undergoing the greatest development and improvement as the results of this stage significantly affect the final classification of objects and the interpretation of input information. Typically, computer vision systems use convolutional neural networks for training (Convolution Neural Network, CNN). The disadvantages of such systems are significant limitations in cross-modal learning, multimodality implementation, labeling of large amounts of data, etc. One of the ways to overcome these problems is to use Vision Transformers (ViT), which, compared to classical CNNs, have higher performance due to reduced inductive biases and high parallel computing efficiency. Introducing Self-Supervised Learning (SSL) technologies can significantly reduce the dependence on manually labeled data, contributing to the formation of generalized representations of images. Cross-Modal Learning (CML) expands the possibilities of processing them by combining data of different types. The development of the new approach, combined with the capabilities of cross-modal learning and self-learning in ViT in a single architecture, will ensure adaptability, efficiency, and system scalability in various applications. The research aims to provide a detailed overview of ViTs, approaches to their architecture, and methods for improving their efficiency. The mathematical foundations of the key concepts of ViT, cross-modal learning and self-learning, the main modifications of ViT, and their integration with SSL and CML technologies are considered. A comparison of methods using characteristics, performance, and efficiency is provided. The key challenges and prospects facing researchers and developers while creating universal models in computer vision are outlined. ViTs change computer vision by capturing global dependencies on images. Despite some challenges, ViTs provide excellent scalability and performance for large datasets. The active search for methods to overcome their limitations makes ViTs a key tool for improving image classification, object detection, and other computer vision tasks.

Fairness has become an important agenda in computer vision and artificial intelligence. Recent studies have shown that many computer vision models and datasets exhibit demographic biases and proposed mitigation strategies. These works attempt to address accuracy disparity, spurious correlations, or unbalanced representations in datasets in tasks such as face recognition, verification and expression and attribute classification. These tasks, however, all require face detection as the first preprocessing step, and surprisingly, there has been little effort in identifying or mitigating biases in face detection. Biased face detectors themselves pose a threat against fair and ethical AI systems, and their biases may be further passed on to subsequent downstream tasks such as face recognition in a computer vision pipeline. This paper therefore investigates the problem of biases in face detection, focusing on accuracy disparity of detectors between demographic groups including gender, age group, and skin tone. We collect perceived demographic attributes on a popular face detection benchmark dataset, WIDER FACE, report skewed demographic distributions, and compare detection performance between groups. In order to mitigate the biases, we apply three mitigation methods that have been introduced in the recent literature and also propose two novel methods. Experimental results show that these methods are effective in reducing demographic biases. We also discuss how the effectiveness varies by demographic attributes, detection easiness, and multiple detectors, which will shed light on this new topic of addressing face detection bias.

More and more businesses are using computer vision systems, which has made it clear that they need strong defenses against threats from other companies. These attacks, which change raw data in small ways, can lead to big mistakes in classification, which makes AI-driven systems less reliable. There have been recent improvements in defense mechanisms for computer vision models against risks from other computers. This study paper looks at both old and new methods. The outcome documented in this study paper shows that different protection strategies work. Different levels of defense are provided by traditional techniques such as adversarial training, input preparation, and defensive distilling. Of these, adversarial training is the most effective against known threats. A lot of the time, these methods have trouble with new or complex hostile changes. More advanced methods, like verified stability and feature squeezing, work better, especially against more complicated threats. The most accurate defenses are hybrids that use antagonistic training along with feature manipulation or GAN-based cleaning. These methods offer a fuller defense. In the world of computer vision, the fight between attackers and defenders is still going on, and new attack methods are always coming out. New improvements in security systems have made computer vision models more durable, but there is still a lot of work to be done. In the future, hybrid and adaptable barriers could be very useful because they offer more flexible and strong security. To keep AI systems safe and reliable, these defenses must keep getting better. This is especially important for important uses like self-driving cars, healthcare, and spying.

The papers in this special section address the programs and services supported by graphical models in computer vision. This section explores the main challenges in this framework—modeling novel priors, learning, inference—and presents innovative solutions. The papers cover the aspects of modeling novel priors, inference algorithms and parameter learning methods in the context of higher order graphical models.

DeepSolar++: Understanding residential solar adoption trajectories with computer vision and technology diffusion models

This comprehensive volume is an essential reference tool for professional and academic researchers in the filed of computer vision, image processing, and applied mathematics. Continuing rapid advances in image processing have been enhanced by the theoretical efforts of mathematicians and engineers. This marriage of mathematics and computer vision - computational vision - has resulted in a discrete approach to image processing that is more reliable when leveraging in practical tasks. This comprehensive volume provides a detailed discourse on the mathematical models used in computational vision from leading educators and active research experts in this field. Topical areas include: image reconstruction, segmentation and object extraction, shape modeling and registration, motion analysis and tracking, and 3D from images, geometry and reconstruction. The book also includes a study of applications in medical image analysis. Handbook of Mathematical Models in Computer Vision provides a graduate-level treatment of this subject as well as serving as a complete reference work for professionals.

This study evaluates the generalization capabilities of computer vision (CV) models for detecting cow locations in indoor free-stall barns, with a focus on quantifying the resources required for effective deployment across different farms where camera angles and lighting conditions vary. Given the increasing reliance of precision livestock farming (PLF) on object localization techniques to monitor livestock health and optimize resource management, understanding how to deploy CV models with minimal resources while maintaining high performance is critical. To assess generalization, we investigated three key hypotheses: (1) Model generalization is equally influenced by changes in lighting conditions and camera angles; (2) Higher model complexity guarantees better generalization performance; and (3) Fine-tuning with custom initial weights trained on related tasks enhances final performance compared to using pre-trained weights. The study utilized the newly released COws LOcalization (COLO) dataset, comprising 1,254 images and 11,818 cow instances captured from five distinct imaging settings with varying camera angles and lighting conditions. The first hypothesis was tested by training models on images from specific camera angles and evaluating performance on new angles. The second hypothesis was examined by comparing the performance of five different model sizes (e.g., YOLO11n with 2.6M parameters vs. YOLO11x with 56.9M parameters) on the same dataset. The third hypothesis was evaluated by pre-training a model on a subset of images from a specific barn and fine-tuning it on another subset from a similar setting to simulate transferring the training outcome to a different location within the same barn. Results indicate that changes in lighting conditions have a negligible impact on model generalization, whereas shifting from a top-view to a side-view perspective significantly reduces performance (P < 0.001). Model complexity does not always correlate with improved generalization, as optimal complexity depends on the dataset and task. Additionally, fine-tuning with custom initial weights benefits small, simple models but provides no significant (P > 0.05) advantage for larger models compared to training from scratch. As artificial intelligence adoption grows, alongside rising concerns about its environmental impact, this study provides practical guidelines for PLF researchers to optimize CV model deployment with minimal resource consumption. It also highlights key generalization challenges and leverages the COLO dataset to drive further research in resource-efficient livestock monitoring.

Applying advanced technologies such as computer vision is highly desirable in seed testing. Among testing needs, computer vision is a feasible technology for conducting seed and seedling classification used in purity analysis and in germination tests. This review focuses on seed identification that currently encounters extreme challenges due to a shortage of expertise, time-consuming training and operation, and the need for large numbers of reference specimens. The reviewed computer vision techniques and application strategies also apply to other methods in seed testing. The review describes the development of machine learning-based computer vision in automating seed identification and their limitations in feature extraction and accuracy. As a subset of machine learning techniques, deep learning has been applied successfully in many agricultural domains, which presents potential opportunities for its application in seed identification and seed testing. To facilitate application in seed testing, the challenges of deep learning-based computer vision systems are summarised through analysing their application in other agricultural domains. It is recommended to accelerate the application in seed testing by optimising procedures or approaches in image acquisition technologies, dataset construction and model development. A concept flow chart for using computer vision systems is proposed to advance computer-assisted seed identification.

Computer Vision, and hence Artificial Intelligence-based extraction of information from images, has increasingly received attention over the last years, for instance in medical diagnostics. While the algorithms' complexity is a reason for their increased performance, it also leads to the "black box" problem, consequently decreasing trust towards AI. In this regard, "Explainable Artificial Intelligence" (XAI) allows to open that black box and to improve the degree of AI transparency. In this paper, we first discuss the theoretical impact of explainability on trust towards AI, followed by showcasing how the usage of XAI in a health-related setting can look like. More specifically, we show how XAI can be applied to understand why Computer Vision, based on deep learning, did or did not detect a disease (malaria) on image data (thin blood smear slide images). Furthermore, we investigate, how XAI can be used to compare the detection strategy of two different deep learning models often used for Computer Vision: Convolutional Neural Network and Multi-Layer Perceptron. Our empirical results show that i) the AI sometimes used questionable or irrelevant data features of an image to detect malaria (even if correctly predicted), and ii) that there may be significant discrepancies in how different deep learning models explain the same prediction. Our theoretical discussion highlights that XAI can support trust in Computer Vision systems, and AI systems in general, especially through an increased understandability and predictability.

The unprecedented acceleration in the adoption and development of artificial intelligence (AI) applications-spanning domains such as natural language processing, computer vision, recommendation systems, and generative modeling-

This data article introduces a structured pavement surface image dataset developed to advance research in automated pavement condition assessment and data-driven road infrastructure monitoring. The dataset comprises three distinct pavement condition categories: (i) alligator cracking, (ii) edge-breaking distress, and (iii) undamaged (intact) pavement surfaces, each representing a prevalent form of pavement deterioration or intact condition typically observed in flexible pavements. The dataset consists of 12,000 raw images (4000 per class) collected under real-world conditions. These images represent the primary scientific contribution of the dataset. All images were standardized through resizing and normalization, and the dataset was partitioned into training, validation, and testing subsets to ensure reproducibility and consistency in data-driven experiments. Pavement images were collected from selected segments of National Highway N6 in Pabna District, Bangladesh, under natural daylight conditions using a smartphone camera during field surveys. Image acquisition was conducted following standard safety practices without disrupting traffic flow. All images were manually reviewed and labelled to ensure annotation accuracy. This dataset is intended to support research on automated pavement crack detection and classification, benchmarking of computer vision and deep learning models, and the development of lightweight and edge-deployable inspection systems.

THINGS-data reflects three large-scale neuroimaging and behavioral datasets of object processing in humans, comprising densely sampled functional MRI and magnetoencephalographic recordings, as well as 4.70 million similarity judgments in response to thousands of photographic images for up to 1854 objects.

Colorectal cancer (CRC) is the third most diagnosed cancer form globally and the second leading cause of cancer-related death after lung cancer. A precise histological categorization of CRC tissue is critical for diagnosis and patient management decisions. However, the variety of tissue patterns in CRC histological images makes the classification a challenging problem. This study applies Vision Transformers, a new class of deep-learning models in computer vision, to perform a multiclass tissue classification of a publicly available CRC histology images dataset. The data set consists of 5000 images with eight categories of tissue. We trained two variants of Transformers, namely Vision Transformer and Compact Convolutional Transformer, and achieved 93.3% and 95% accuracy, respectively. Our results outperform the original paper (87.4%) on the same dataset. Furthermore, our study highlights the opportunities of using Transformers in the histopathological image domain.

The development of PACS image databases has long been thought of as a major technological challenge, due to the amount of data to be managed. On the contrary, the authors think that despite major improvements in storage technology, new data management techniques must be proposed to make image databases medically useful in PACS environments. More precisely, image databases must contain not only images per se, but also the description of all objects used in medical imaging, in order to permit the remote processing, analysis, and interpretation of images. In several other papers, the authors explain why they adopted an object-oriented approach to model information in medical imaging. In this paper, the focus is on the inventory of objects manipulated in medical imaging from a qualitative viewpoint. For this purpose, a large number of representative imaging procedures were selected. The authors characterized how they are asked for by clinicians, realized in imaging departments, and consumed by requiring physicians and surgeons, in three French university hospitals. On the basis of this inventory, a set of image data -- i.e. of objects used in medical imaging -- was defined to show that this set must evolve with advances in medical imaging, and to point out that relational DBMS concepts cannot represent all image data.