The image deepfake detection task has been greatly addressed by the scientific community to discriminate real images from those generated by Artificial Intelligence (AI) models: a binary classification task. In this work, the deepfake detection and recognition task was investigated by collecting a dedicated dataset of pristine images and fake ones generated by 9 different Generative Adversarial Network (GAN) architectures and by 4 additional Diffusion Models (DM). A hierarchical multi-level approach was then introduced to solve three different deepfake detection and recognition tasks: (i) Real Vs AI generated; (ii) GANs Vs DMs; (iii) AI specific architecture recognition. Experimental results demonstrated, in each case, more than 97% classification accuracy, outperforming state-of-the-art methods.
Image fusion in Remote Sensing (RS) has been a consistent demand due to its ability to turn raw images of different resolutions, sources, and modalities into accurate, complete, and spatio-temporally coherent images. It greatly facilitates downstream applications such as pan-sharpening, change detection, land-cover classification, etc. Yet, image fusion solutions are highly disparate to various remote sensing problems and thus are often narrowly defined in existing reviews as topical applications, such as pan-sharpening, and spatial-temporal image fusion. Considering that image fusion can be theoretically applied to any gridded data through pixel-level operations, in this paper, we expanded its scope by comprehensively surveying relevant works with a simple taxonomy: 1) many-to-one image fusion; 2) many-to-many image fusion. This simple taxonomy defines image fusion as a mapping problem that turns either a single or a set of images into another single or set of images, depending on the desired coherence, e.g., spectral, spatial/resolution coherence, etc. We show that this simple taxonomy, despite the significant modality difference it covers, can be presented by a conceptually easy framework. In addition, we provide a meta-analysis to review the major papers studying the various types of image fusion and their applications over the years (from the 1980s to date), covering 5,926 peer-reviewed papers. Finally, we discuss the main benefits and emerging challenges to provide open research directions and potential future works.
Medical image analysis increasingly relies on the integration of multiple imaging modalities to capture complementary anatomical and functional information, enabling more accurate diagnosis and treatment planning. Achieving aligned feature representations across these diverse modalities is therefore important for effective multimodal analysis. While contrastive language-image pre-training (CLIP) and its variant have enabled image-text alignments, they require explicitly paired data between arbitrary two modalities, which is difficult to acquire in medical contexts. To address the gap, we present Multimodal Medical Image Binding with Text (M\textsuperscript{3}Bind), a novel pre-training framework that enables seamless alignment of multiple medical imaging modalities through a shared text representation space without requiring explicit paired data between any two medical image modalities. Specifically, based on the insight that different images can naturally bind with text, M\textsuperscript{3}Bind first fine-tunes pre-trained CLIP-like image-text models to align their modality-specific text embedding space while preserving their original image-text alignments. Subsequently, we distill these modality-specific text encoders into a unified model, creating a shared text embedding space. Experiments on X-ray, CT, retina, ECG, and pathological images on multiple downstream tasks demonstrate that M\textsuperscript{3}Bind achieves state-of-the-art performance in zero-shot, few-shot classification and cross-modal retrieval tasks compared to its CLIP-like counterparts. These results validate M\textsuperscript{3}Bind's effectiveness in achieving cross-image-modal alignment for medical analysis.
This paper proposes a self-explainable Deep Learning (SE-DL) system for an image classification problem that performs self-error detection. The self-error detection is key to improving the DL system's safe operation, especially in safety-critical applications such as automotive systems. A SE-DL system outputs both the class prediction and an explanation for that prediction, which provides insight into how the system makes its predictions for humans. Additionally, we leverage the explanation of the proposed SE-DL system to detect potential class prediction errors of the system. The proposed SE-DL system uses a set of concepts to generate the explanation. The concepts are human-understandable lower-level image features in each input image relevant to the higher-level class of that image. We present a concept selection methodology for scoring all concepts and selecting a subset of them based on their contribution to the error detection performance of the proposed SE-DL system. Finally, we present different error detection schemes using the proposed SE-DL system to compare them against an error detection scheme without any SE-DL system.
Genetic Programming (GP) has been primarily used to tackle optimization, classification, and feature selection related tasks. The widespread use of GP is due to its flexible and comprehensible tree-type structure. Similarly, research is also gaining momentum in the field of Image Processing, because of its promising results over vast areas of applications ranging from medical Image Processing to multispectral imaging. Image Processing is mainly involved in applications such as computer vision, pattern recognition, image compression, storage, and medical diagnostics. This universal nature of images and their associated algorithm, i.e., complexities, gave an impetus to the exploration of GP. GP has thus been used in different ways for Image Processing since its inception. Many interesting GP techniques have been developed and employed in the field of Image Processing, and consequently, we aim to provide the research community an extensive view of these techniques. This survey thus presents the diverse applications of GP in Image Processing and provides useful resources for further research. Also, the comparison of different parameters used in different applications of Image Processing is summarized in tabular form. Moreover, analysis of the different parameters used in Image Processing related tasks is carried-out to save the time needed in the future for evaluating the parameters of GP. As more advancement is made in GP methodologies, its success in solving complex tasks, not only in Image Processing but also in other fields, may increase. Additionally, guidelines are provided for applying GP in Image Processing related tasks, the pros and cons of GP techniques are discussed, and some future directions are also set.
Quantum Channel Discrimination (QCD) presents a fundamental task in quantum information theory, with critical applications in quantum reading, illumination, data-readout and more. The extension to multiple quantum channel discrimination has seen a recent focus to characterise potential quantum advantage associated with quantum enhanced discriminatory protocols. In this paper, we study thermal imaging as an environment localisation task, in which thermal images are modelled as ensembles of Gaussian phase insensitive channels with identical transmissivity, and pixels possess properties according to background (cold) or target (warm) thermal channels. Via the teleportation stretching of adaptive quantum protocols, we derive ultimate limits on the precision of pattern classification of abstract, binary thermal image spaces, and show that quantum enhanced strategies may be used to provide significant quantum advantage over known optimal classical strategies. The environmental conditions and necessary resources for which advantage may be obtained are studied and discussed. We then numerically investigate the use of quantum enhanced statistical classifiers, in which quantum sensors are used in conjunction with machine learning image classification methods. Proving definitive advantage in the low loss regime, this work motivates the use of quantum enhanced sources for short-range thermal imaging and detection techniques for future quantum technologies.
Highly imbalanced datasets are ubiquitous in medical image classification problems. In such problems, it is often the case that rare classes associated to less prevalent diseases are severely under-represented in labeled databases, typically resulting in poor performance of machine learning algorithms due to overfitting in the learning process. In this paper, we propose a novel mechanism for sampling training data based on the popular MixUp regularization technique, which we refer to as Balanced-MixUp. In short, Balanced-MixUp simultaneously performs regular (i.e., instance-based) and balanced (i.e., class-based) sampling of the training data. The resulting two sets of samples are then mixed-up to create a more balanced training distribution from which a neural network can effectively learn without incurring in heavily under-fitting the minority classes. We experiment with a highly imbalanced dataset of retinal images (55K samples, 5 classes) and a long-tail dataset of gastro-intestinal video frames (10K images, 23 classes), using two CNNs of varying representation capabilities. Experimental results demonstrate that applying Balanced-MixUp outperforms other conventional sampling schemes and loss functions specifically designed to deal with imbalanced data. Code is released at https://github.com/agaldran/balanced_mixup .
Recent years have seen a growth in user-centric applications that require effective knowledge transfer across tasks in the low-data regime. An example is personalization, where a pretrained system is adapted by learning on small amounts of labeled data belonging to a specific user. This setting requires high accuracy under low computational complexity, therefore the Pareto frontier of accuracy vs. adaptation cost plays a crucial role. In this paper we push this Pareto frontier in the few-shot image classification setting with a key contribution: a new adaptive block called Contextual Squeeze-and-Excitation (CaSE) that adjusts a pretrained neural network on a new task to significantly improve performance with a single forward pass of the user data (context). We use meta-trained CaSE blocks to conditionally adapt the body of a network and a fine-tuning routine to adapt a linear head, defining a method called UpperCaSE. UpperCaSE achieves a new state-of-the-art accuracy relative to meta-learners on the 26 datasets of VTAB+MD and on a challenging real-world personalization benchmark (ORBIT), narrowing the gap with leading fine-tuning methods with the benefit of orders of magnitude lower adaptation cost.
Zero-shot incremental learning aims to enable the model to generalize to new classes without forgetting previously learned classes. However, the semantic gap between old and new sample classes can lead to catastrophic forgetting. Additionally, existing algorithms lack capturing significant information from each sample image domain, impairing models' classification performance. Therefore, this paper proposes a novel Spatial-Frequency Domain Network (SFDNet) which contains a Spatial-Frequency Feature Extraction (SFFE) module and Attention Feature Alignment (AFA) module to improve the Zero-Shot Translation for Class Incremental algorithm. Firstly, SFFE module is designed which contains a dual attention mechanism for obtaining salient spatial-frequency feature information. Secondly, a novel feature fusion module is conducted for obtaining fused spatial-frequency domain features. Thirdly, the Nearest Class Mean classifier is utilized to select the most suitable category. Finally, iteration between tasks is performed using the Zero-Shot Translation model. The proposed SFDNet has the ability to effectively extract spatial-frequency feature representation from input images, improve the accuracy of image classification, and fundamentally alleviate catastrophic forgetting. Extensive experiments on the CUB 200-2011 and CIFAR100 datasets demonstrate that our proposed algorithm outperforms state-of-the-art incremental learning algorithms.
The recent emergence of the hybrid quantum-classical neural network (HQCNN) architecture has garnered considerable attention due to the potential advantages associated with integrating quantum principles to enhance various facets of machine learning algorithms and computations. However, the current investigated serial structure of HQCNN, wherein information sequentially passes from one network to another, often imposes limitations on the trainability and expressivity of the network. In this study, we introduce a novel architecture termed Parallel Proportional Fusion of Quantum and Spiking Neural Networks (PPF-QSNN). The dataset information is simultaneously fed into both the spiking neural network and the variational quantum circuits, with the outputs amalgamated in proportion to their individual contributions. We systematically assess the impact of diverse PPF-QSNN parameters on network performance for image classification, aiming to identify the optimal configuration. Numerical results on the MNIST dataset unequivocally illustrate that our proposed PPF-QSNN outperforms both the existing spiking neural network and the serial quantum neural network across metrics such as accuracy, loss, and robustness. This study introduces a novel and effective amalgamation approach for HQCNN, thereby laying the groundwork for the advancement and application of quantum advantage in artificial intelligent computations.
Transformers have become the de facto standard for a wide range of tasks, from image classification to physics simulations. Despite their impressive performance, the quadratic complexity of standard Transformers in both memory and time with respect to the input length makes them impractical for processing high-resolution inputs. Therefore, several variants have been proposed, the most successful relying on patchification, downsampling, or coarsening techniques, often at the cost of losing the finest-scale details. In this work, we take a different approach. Inspired by state-of-the-art techniques in $n$-body numerical simulations, we cast attention as an interaction problem between grid points. We introduce the Multipole Attention Neural Operator (MANO), which computes attention in a distance-based multiscale fashion. MANO maintains, in each attention head, a global receptive field and achieves linear time and memory complexity with respect to the number of grid points. Empirical results on image classification and Darcy flows demonstrate that MANO rivals state-of-the-art models such as ViT and Swin Transformer, while reducing runtime and peak memory usage by orders of magnitude. We open source our code for reproducibility at https://github.com/AlexColagrande/MANO.
Driven by the high profit, Portable Executable (PE) malware has been consistently evolving in terms of both volume and sophistication. PE malware family classification has gained great attention and a large number of approaches have been proposed. With the rapid development of machine learning techniques and the exciting results they achieved on various tasks, machine learning algorithms have also gained popularity in the PE malware family classification task. Three mainstream approaches that use learning based algorithms, as categorized by the input format the methods take, are image-based, binary-based and disassembly-based approaches. Although a large number of approaches are published, there is no consistent comparisons on those approaches, especially from the practical industry adoption perspective. Moreover, there is no comparison in the scenario of concept drift, which is a fact for the malware classification task due to the fast evolving nature of malware. In this work, we conduct a thorough empirical study on learning-based PE malware classification approaches on 4 different datasets and consistent experiment settings. Based on the experiment results and an interview with our industry partners, we find that (1) there is no individual class of methods that significantly outperforms the others; (2) All classes of methods show performance degradation on concept drift (by an average F1-score of 32.23%); and (3) the prediction time and high memory consumption hinder existing approaches from being adopted for industry usage.
Remote sensing scene classification (RSSC) is traditionally formulated as a vision-centric problem, where inference relies on rich visual representations extracted from images. In this work, we revisit this assumption by introducing text-only RSSC, a new problem setting that investigates whether remote sensing scenes can be recognized using only textual semantic representations. To enable this study, we propose SeProRS, which reformulates RSSC as a text-only classification task through semantic proxy generation. Specifically, a vision-language model (VLM) is first employed to convert remote sensing images into textual descriptions, termed semantic proxies, which are then encoded by a text embedding model for scene classification. This decoupled pipeline provides a unified and reproducible benchmark for evaluating text-only scene understanding. Furthermore, we investigate representative prompting strategies and proxy generators, and analyze their impact on classification performance, text embedding behavior, cross-generator transferability, and efficiency. Extensive experiments on three well-known RSSC datasets show that semantic proxies can preserve substantial discriminative information, while also revealing inherent limitations of purely text-based scene understanding. This paper does not aim to replace image-based RSSC, but instead establishes a new research direction by quantifying the feasibility and boundaries of semantic proxy-based scene recognition. The proposed benchmark and analysis provide a foundation for future research on efficient and modality-reduced remote sensing scene understanding. The source code and datasets will be accessible at https://github.com/sstary/SSRS.
An automatic method for the selection of subsets of images, both modern and historic, out of a set of landmark large images collected from the Internet is presented in this paper. This selection depends on the extraction of dominant features using Gabor filtering. Features are selected carefully from a preliminary image set and fed into a neural network as a training data. The method collects a large set of raw landmark images containing modern and historic landmark images and non-landmark images. The method then processes these images to classify them as landmark and non-landmark images. The classification performance highly depends on the number of candidate features of the landmark.
This work tackles the problem of semi-supervised learning of image classifiers. Our main insight is that the field of semi-supervised learning can benefit from the quickly advancing field of self-supervised visual representation learning. Unifying these two approaches, we propose the framework of self-supervised semi-supervised learning and use it to derive two novel semi-supervised image classification methods. We demonstrate the effectiveness of these methods in comparison to both carefully tuned baselines, and existing semi-supervised learning methods. We then show that our approach and existing semi-supervised methods can be jointly trained, yielding a new state-of-the-art result on semi-supervised ILSVRC-2012 with 10% of labels.
We show that the basic classification framework alone can be used to tackle some of the most challenging tasks in image synthesis. In contrast to other state-of-the-art approaches, the toolkit we develop is rather minimal: it uses a single, off-the-shelf classifier for all these tasks. The crux of our approach is that we train this classifier to be adversarially robust. It turns out that adversarial robustness is precisely what we need to directly manipulate salient features of the input. Overall, our findings demonstrate the utility of robustness in the broader machine learning context. Code and models for our experiments can be found at https://git.io/robust-apps.
We investigate the problem of training neural networks from incomplete images without replacing missing values. For this purpose, we first represent an image as a graph, in which missing pixels are entirely ignored. The graph image representation is processed using a spatial graph convolutional network (SGCN) -- a type of graph convolutional networks, which is a proper generalization of classical CNNs operating on images. On one hand, our approach avoids the problem of missing data imputation while, on the other hand, there is a natural correspondence between CNNs and SGCN. Experiments confirm that our approach performs better than analogical CNNs with the imputation of missing values on typical classification and reconstruction tasks.
Meta-learning algorithms are widely used for few-shot learning. For example, image recognition systems that readily adapt to unseen classes after seeing only a few labeled examples. Despite their success, we show that modern meta-learning algorithms are extremely sensitive to the data used for adaptation, i.e. support data. In particular, we demonstrate the existence of (unaltered, in-distribution, natural) images that, when used for adaptation, yield accuracy as low as 4\% or as high as 95\% on standard few-shot image classification benchmarks. We explain our empirical findings in terms of class margins, which in turn suggests that robust and safe meta-learning requires larger margins than supervised learning.
The Gaussian kernel and its derivatives have already been employed for Convolutional Neural Networks in several previous works. Most of these papers proposed to compute filters by linearly combining one or several bases of fixed or slightly trainable Gaussian kernels with or without their derivatives. In this article, we propose a high-level configurable layer based on anisotropic, oriented and shifted Gaussian derivative kernels which generalize notions encountered in previous related works while keeping their main advantage. The results show that the proposed layer has competitive performance compared to previous works and that it can be successfully included in common deep architectures such as VGG16 for image classification and U-net for image segmentation.
We explore an efficient approach to establish a foundational video-text model. We present VideoCoCa that maximally reuses a pretrained image-text contrastive captioner (CoCa) model and adapt it to video-text tasks with minimal extra training. While previous works adapt image-text models with various cross-frame fusion modules, we find that the generative attentional pooling and contrastive attentional pooling layers in CoCa are instantly adaptable to flattened frame embeddings, yielding state-of-the-art results on zero-shot video classification and zero-shot text-to-video retrieval. Furthermore, we explore lightweight finetuning on top of VideoCoCa, and achieve strong results on video question-answering and video captioning.