Prompt-based image segmentation has revolutionized computer vision by enabling more adaptive and efficient segmentation through prompts. In the context of image segmentation, the term prompt broadly refers to any auxiliary input, such as clicks, boxes, scribbles, support sets or free-form text that guides a model’s segmentation behavior. These inputs operate as task-specific signals that enable models to adapt their segmentation behavior to different contexts and objectives. This survey categorizes promptable image segmentation into five primary areas: interactive segmentation, referring segmentation, few-shot semantic segmentation, open vocabulary segmentation and foundation models. The authors explore how different prompting strategies improve segmentation performance while enabling few-shot learning and reducing reliance on extensive labeled datasets. The discussion highlights the role of foundation models in advancing segmentation capabilities by integrating separate components of these complex models and leveraging multimodal interactions. By synthesizing state-of-the-art techniques, this study provides a structured taxonomy, identifies key challenges in multimodal fusion and generalization and outlines future directions for developing more intelligent and adaptable segmentation systems.
Sliced Optimal Transport (SOT) is a rapidly developing branch of optimal transport (OT) that exploits the tractability of one-dimensional OT problems. By combining tools from OT, integral geometry, and computational statistics, SOT enables fast and scalable computation of distances, barycenters, and kernels for probability measures, while retaining rich geometric structure. This work provides a comprehensive review of SOT, covering its mathematical foundations, methodological advances, computational methods, and applications. We discuss key concepts of OT and one-dimensional OT, the role of tools from integral geometry such as Radon transform in projecting measures, and statistical techniques for estimating sliced distances. The work further explores recent methodological advances, including non-linear projections, improved Monte Carlo approximations, statistical estimation techniques for one-dimensional optimal transport, weighted slicing techniques, and transportation plan estimation methods. Variational problems, such as minimum sliced Wasserstein estimation, barycenters, gradient flows, kernel constructions, and embeddings are examined alongside extensions to unbalanced, partial, multi-marginal, and Gromov-Wasserstein settings. Applications span machine learning, statistics, computer graphics and computer visions, highlighting SOT's versatility as a practical computational tool. This work will be of interest to researchers and practitioners in machine learning, data sciences, and computational disciplines seeking efficient alternatives to classical OT.
We present an accessible first course on the mathematics of diffusion models and flow matching for machine learning. We aim to teach diffusion as simply as possible, with minimal mathematical and machine learning prerequisites, but enough technical detail to reason about its correctness. Unlike most tutorials on this subject, we take neither a Variational Auto Encoder (VAE) nor a Stochastic Differential Equations (SDE) approach. In fact, for the core ideas we will not need any SDEs, Evidence-Based-Lower-Bounds (ELBOs), Langevin dynamics, or even the notion of a score. The reader need only be familiar with basic probability, calculus, linear algebra, and multivariate Gaussians. This tutorial has five parts, each relatively self-contained. Section 1 presents the fundamentals of diffusion: the problem we are trying to solve and an overview of the basic approach. Sections 2 and 3 show how to construct a stochastic and deterministic diffusion sampler, respectively, and give intuitive derivations for why these samplers correctly reverse the forward diffusion process. Section 4 covers the closely-related topic of Flow Matching, which can be thought of as a generalization of diffusion that offers additional flexibility (including what are called rectified flows or linear flows). Finally, in Section 5 we return to diffusion and connect this tutorial to the broader literature while highlighting some of the design choices that matter most in practice, including samplers, noise schedules, and parametrizations.
Despite the growing interest in diffusion models, gaining a deep understanding of the model class remains an elusive endeavour, particularly for the uninitiated in non-equilibrium statistical physics. Thanks to the rapid rate of progress in the field, most existing work on diffusion models focuses on either applications or theoretical contributions. Unfortunately, the theoretical material is often inaccessible to practitioners and new researchers, leading to a risk of superficial understanding in ongoing research. Given that diffusion models are now an indispensable tool, a clear and consolidating perspective on the model class is needed to properly contextualize recent advances in generative modelling and lower the barrier to entry for new researchers. To that end, we revisit predecessors to diffusion models, such as hierarchical latent variable models, and synthesize a holistic perspective using only directed graphical modelling and variational inference principles. The resulting narrative is easier to follow as it imposes fewer prerequisites on the average reader relative to the view from non-equilibrium thermodynamics or stochastic differential equations.
The field of computer vision is now a multi-billion dollar enterprise, with its use in surveillance applications driving this large market share. In the last six years, computer vision researchers have started to discuss the risks and harms of some of these systems, mostly using the lens of fairness introduced in the machine learning literature to perform this analysis. While this lens is useful to uncover and mitigate a narrow segment of the harms that can be enacted through computer vision systems, it is only one of the toolkits that researchers have available to uncover and mitigate the harms of the systems they build. In this monograph, we discuss a wide range of risks and harms that can be enacted through the development and deployment of computer vision systems. We also discuss some existing technical approaches to mitigating these harms, as well as the shortcomings of these mitigation strategies. Then, we introduce computer vision researchers to harm mitigation strategies proposed by journalists, human rights activists, individuals harmed by computer vision systems, and researchers in disciplines ranging from sociology to physics. We conclude the monograph by listing principles that researchers can follow to build what we call community-rooted computer vision tools in the public interest, and give examples of such research directions. We hope that this monograph can serve as a starting point for researchers exploring the harms of current computer vision systems and attempting to steer the field into community-rooted work.
This paper presents a comprehensive survey of the taxonomy and evolution of multimodal foundation models that demonstrate vision and vision-language capabilities, focusing on the transition from specialist models to general-purpose assistants. The research landscape encompasses five core topics, categorized into two classes. (i) We start with a survey of well-established research areas: multimodal foundation models pre-trained for specific purposes, including two topics -- methods of learning vision backbones for visual understanding and text-to-image generation. (ii) Then, we present recent advances in exploratory, open research areas: multimodal foundation models that aim to play the role of general-purpose assistants, including three topics -- unified vision models inspired by large language models (LLMs), end-to-end training of multimodal LLMs, and chaining multimodal tools with LLMs. The target audiences of the paper are researchers, graduate students, and professionals in computer vision and vision-language multimodal communities who are eager to learn the basics and recent advances in multimodal foundation models.
The astonishing growth of generative tools in recent years has empowered many exciting applications in text-to-image generation and text-to-video generation. The underlying principle behind these generative tools is the concept of diffusion, a particular sampling mechanism that has overcome some shortcomings that were deemed difficult in the previous approaches. The goal of this tutorial is to discuss the essential ideas underlying the diffusion models. The target audience of this tutorial includes undergraduate and graduate students who are interested in doing research on diffusion models or applying these models to solve other problems.
Since the seminal work of Andrey Kolmogorov in the early 1940's, imaging through atmospheric turbulence has grown from a pure scientific pursuit to an important subject across a multitude of civilian, space-mission, and national security applications. Fueled by the recent advancement of deep learning, the field is further experiencing a new wave of momentum of applying these learning-based techniques to the problem. However, because of the complexity of the physics of atmospheric turbulence, significant gaps remain to be filled before the power of deep learning can be fully unleashed. In particular, the goal of building the most accurate turbulence model to mimic nature is gradually shifted to designing a compromised model that can maximize the image reconstruction performance. This leads to a new field which this book is trying to explain, Computational Imaging Through Atmospheric Turbulence. The goal of this book is to present the basic concepts of turbulence physics while framing it under the theme of computational imaging. Emphasis is put on elaborating the principles of how waves propagate through atmospheric turbulence and propagation-free approaches to reproduce the effect without needing wave propagation equations. This allows for a much faster simulation while preserving the physics of turbulence, hence creating the possibility of integrating turbulence physics into the design of image reconstruction algorithms. The book is written for readers with an image processing background who are seeking to understand the physics of turbulence. Connections with deep learning are emphasized throughout the book.
Visual compression is an application of data compression to lower the storage and/or transmission requirements for digital images and videos. Due to the rapid growth in visual data transmission demand, more efficient compression algorithms are needed. Considering that deep learning techniques have successfully revolutionized many visual tasks, learning-based compression algorithms have been explored over the years and have been shown to be able to outperform many conventional compression methods. This survey provides a review of various visual compression algorithms, both end-to-end learning-based image compression approaches and hybrid image compression approaches. Some learningbased video compression methods are also discussed. In addition to describing a wide range of learning-based image compression approaches that have been developed in recent years, the survey describes widely used datasets, presents recent standardization efforts, and discusses potential research directions.
Online crowdsourcing platforms have made it increasingly easy to perform evaluations of algorithm outputs with survey questions like "which image is better, A or B?", leading to their proliferation in vision and graphics research papers. Results of these studies are often used as quantitative evidence in support of a paper's contributions. On the one hand we argue that, when conducted hastily as an afterthought, such studies lead to an increase of uninformative, and, potentially, misleading conclusions. On the other hand, in these same communities, user research is underutilized in driving project direction and forecasting user needs and reception. We call for increased attention to both the design and reporting of user studies in computer vision and graphics papers towards (1) improved replicability and (2) improved project direction. Together with this call, we offer an overview of methodologies from user experience research (UXR), human-computer interaction (HCI), and applied perception to increase exposure to the available methodologies and best practices. We discuss foundational user research methods (e.g., needfinding) that are presently underutilized in computer vision and graphics research, but can provide valuable project direction. We provide further pointers to the literature for readers interested in exploring other UXR methodologies. Finally, we describe broader open issues and recommendations for the research community.
Neural compression is the application of neural networks and other machine learning methods to data compression. Recent advances in statistical machine learning have opened up new possibilities for data compression, allowing compression algorithms to be learned end-to-end from data using powerful generative models such as normalizing flows, variational autoencoders, diffusion probabilistic models, and generative adversarial networks. The present article aims to introduce this field of research to a broader machine learning audience by reviewing the necessary background in information theory (e.g., entropy coding, rate-distortion theory) and computer vision (e.g., image quality assessment, perceptual metrics), and providing a curated guide through the essential ideas and methods in the literature thus far.
With the broad growth of video capturing devices and applications on the web, it is more demanding to provide desired video content for users efficiently. Video summarization facilitates quickly grasping video content by creating a compact summary of videos. Much effort has been devoted to automatic video summarization, and various problem settings and approaches have been proposed. Our goal is to provide an overview of this field. This survey covers early studies as well as recent approaches which take advantage of deep learning techniques. We describe video summarization approaches and their underlying concepts. We also discuss benchmarks and evaluations. We overview how prior work addressed evaluation and detail the pros and cons of the evaluation protocols. Last but not least, we discuss open challenges in this field.
Image segmentation is the task of associating pixels in an image with their respective object class labels. It has a wide range of applications in many industries including healthcare, transportation, robotics, fashion, home improvement, and tourism. Many deep learning-based approaches have been developed for image-level object recognition and pixel-level scene understanding-with the latter requiring a much denser annotation of scenes with a large set of objects. Extensions of image segmentation tasks include 3D and video segmentation, where units of voxels, point clouds, and video frames are classified into different objects. We use"Object Segmentation"to refer to the union of these segmentation tasks. In this monograph, we investigate both traditional and modern object segmentation approaches, comparing their strengths, weaknesses, and utilities. We examine in detail the wide range of deep learning-based segmentation techniques developed in recent years, provide a review of the widely used datasets and evaluation metrics, and discuss potential future research directions.
Recent advances in neural signal processing led to significant improvements in the performance of learned image/video restoration and super-resolution (SR). An important benefit of data-driven deep learning approaches to image processing is that neural models can be optimized for any differentiable loss function, including perceptual loss functions, leading to perceptual image/video restoration and SR, which cannot be easily handled by traditional model-based methods. We start with a brief problem statement and a short discussion on traditional vs. data-driven solutions. We next review recent advances in neural architectures, such as residual blocks, dense connections, residual-in-residual dense blocks, residual blocks with generative neurons, self-attention and visual transformers. We then discuss loss functions and evaluation (assessment) criteria for image/video restoration and SR, including fidelity (distortion) and perceptual criteria, and the relation between them, where we briefly review the perception vs. distortion trade-off. We can consider learned image/video restoration and SR as learning either a nonlinear regressive mapping from degraded to ideal images based on the universal approximation theorem, or a generative model that captures the probability distribution of ideal images. We first review regressive inference via residual and/or dense convolutional networks (ConvNet). We also show that using a new architecture with residual blocks based on a generative neuron model can outperform classical residual ConvNets in peak-signal-to-noise ratio (PSNR). We next discuss generative inference based on adversarial training, such as SRGAN and ESRGAN, which can reproduce realistic textures, or based on normalizing flow such as SRFlow by optimizing log-likelihood. We then discuss problems in applying supervised training to real-life restoration and SR, including overfitting image priors and overfitting the degradation model seen in the training set. We introduce multiple-model SR and real-world SR (from unpaired training data) formulations to overcome these problems. Integration of traditional model-based methods and deep learning for non-blind restoration/SR is introduced as another solution to model overfitting in supervised learning. In learned video restoration and SR (VSR), we first discuss how to best exploit temporal correlations in video, including sliding temporal window vs. recurrent architectures for propagation, and aligning frames in the pixel domain using optical flow vs. in the feature space using deformable convolutions. We next introduce early fusion with feature-space alignment, employed by the EDVR network, which obtains excellent PSNR performance. However, it is well-known that videos with the highest PSNR may not be the most appealing to humans, since minimizing the mean-square error may result in blurring of details. We then address perceptual optimization of VSR models to obtain natural texture and motion. Although perception-distortion tradeoff has been well studied for images, few works address perceptual VSR. In addition to using perceptual losses, such as MS-SSIM, LPIPS, and/or adversarial training, we also discuss explicit loss functions/criteria to enforce and evaluate temporal consistency. We conclude with a discussion of open problems.
This paper surveys vision-language pre-training (VLP) methods for multimodal intelligence that have been developed in the last few years. We group these approaches into three categories: ($i$) VLP for image-text tasks, such as image captioning, image-text retrieval, visual question answering, and visual grounding; ($ii$) VLP for core computer vision tasks, such as (open-set) image classification, object detection, and segmentation; and ($iii$) VLP for video-text tasks, such as video captioning, video-text retrieval, and video question answering. For each category, we present a comprehensive review of state-of-the-art methods, and discuss the progress that has been made and challenges still being faced, using specific systems and models as case studies. In addition, for each category, we discuss advanced topics being actively explored in the research community, such as big foundation models, unified modeling, in-context few-shot learning, knowledge, robustness, and computer vision in the wild, to name a few.
Semantic image segmentation (SiS) plays a fundamental role in a broad variety of computer vision applications, providing key information for the global understanding of an image. This survey is an effort to summarize two decades of research in the field of SiS, where we propose a literature review of solutions starting from early historical methods followed by an overview of more recent deep learning methods including the latest trend of using transformers. We complement the review by discussing particular cases of the weak supervision and side machine learning techniques that can be used to improve the semantic segmentation such as curriculum, incremental or self-supervised learning. State-of-the-art SiS models rely on a large amount of annotated samples, which are more expensive to obtain than labels for tasks such as image classification. Since unlabeled data is instead significantly cheaper to obtain, it is not surprising that Unsupervised Domain Adaptation (UDA) reached a broad success within the semantic segmentation community. Therefore, a second core contribution of this book is to summarize five years of a rapidly growing field, Domain Adaptation for Semantic Image Segmentation (DASiS) which embraces the importance of semantic segmentation itself and a critical need of adapting segmentation models to new environments. In addition to providing a comprehensive survey on DASiS techniques, we unveil also newer trends such as multi-domain learning, domain generalization, domain incremental learning, test-time adaptation and source-free domain adaptation. Finally, we conclude this survey by describing datasets and benchmarks most widely used in SiS and DASiS and briefly discuss related tasks such as instance and panoptic image segmentation, as well as applications such as medical image segmentation.
In the last two decades, we have witnessed an immense increase in the use of multimedia content on the internet, for multiple applications ranging from the most innocuous to very critical ones. Naturally, this emergence has given rise to many types of threats posed when this content can be manipulated/used for malicious purposes. For example, fake media can be used to drive personal opinions, ruining the image of a public figure, or for criminal activities such as terrorist propaganda and cyberbullying. The research community has of course moved to counter attack these threats by designing manipulation-detection systems based on a variety of techniques, such as signal processing, statistics, and machine learning. This research and practice activity has given rise to the field of multimedia forensics. The success of deep learning in the last decade has led to its use in multimedia forensics as well. In this survey, we look at the latest trends and deep-learning-based techniques introduced to solve three main questions investigated in the field of multimedia forensics. We begin by examining the manipulations of images and videos produced with editing tools, reporting the deep-learning approaches adopted to counter these attacks. Next, we move on to the issue of the source camera model and device identification, as well as the more recent problem of monitoring image and video sharing on social media. Finally, we look at the most recent challenge that has emerged in recent years: recognizing deep-fakes, which we use to describe any content generated using artificial-intelligence techniques; we present the methods that have been introduced to show the existence of traces left in deepfake content and to detect them. For each problem, we also report the most popular metrics and datasets used today.
Recent years have witnessed enormous progress in AI-related fields such as computer vision, machine learning, and autonomous vehicles. As with any rapidly growing field, it becomes increasingly difficult to stay up-to-date or enter the field as a beginner. While several survey papers on particular sub-problems have appeared, no comprehensive survey on problems, datasets, and methods in computer vision for autonomous vehicles has been published. This monograph attempts to narrow this gap by providing a survey on the state-of-the-art datasets and techniques. Our survey includes both the historically most relevant literature as well as the current state of the art on several specific topics, including recognition, reconstruction, motion estimation, tracking, scene understanding, and end-to-end learning for autonomous driving. Towards this goal, we analyze the performance of the state of the art on several challenging benchmarking datasets, including KITTI, MOT, and Cityscapes. Besides, we discuss open problems and current research challenges. To ease accessibility and accommodate missing references, we also provide a website that allows navigating topics as well as methods and provides additional information.
