Emergency events involving fire are potentially harmful, demanding a fast and precise decision making. The use of crowdsourcing image and videos on crisis management systems can aid in these situations by providing more information than verbal/textual descriptions. Due to the usual high volume of data, automatic solutions need to discard non-relevant content without losing relevant information. There are several methods for fire detection on video using color-based models. However, they are not adequate for still image processing, because they can suffer on high false-positive results. These methods also suffer from parameters with little physical meaning, which makes fine tuning a difficult task. In this context, we propose a novel fire detection method for still images that uses classification based on color features combined with texture classification on superpixel regions. Our method uses a reduced number of parameters if compared to previous works, easing the process of fine tuning the method. Results show the effectiveness of our method of reducing false-positives while its precision remains compatible with the state-of-the-art methods.
Convolutional neural networks (CNNs) are extensively beneficial for medical image processing. Medical images are plentiful, but there is a lack of annotated data. Transfer learning is used to solve the problem of lack of labeled data and grants CNNs better training capability. Transfer learning can be used in many different medical applications; however, the model under transfer should have the same size as the original network. Knowledge distillation is recently proposed to transfer the knowledge of a model to another one and can be useful to cover the shortcomings of transfer learning. But some parts of the knowledge may not be distilled by knowledge distillation. In this paper, a novel knowledge distillation using transfer learning is proposed to transfer the whole knowledge of a model to another one. The proposed method can be beneficial and practical for medical image analysis in which a small number of labeled data are available. The proposed process is tested for diabetic retinopathy classification. Simulation results demonstrate that using the proposed method, knowledge of an extensive network can be transferred to a smaller model.
Large scale image classification datasets often contain noisy labels. We take a principled probabilistic approach to modelling input-dependent, also known as heteroscedastic, label noise in these datasets. We place a multivariate Normal distributed latent variable on the final hidden layer of a neural network classifier. The covariance matrix of this latent variable, models the aleatoric uncertainty due to label noise. We demonstrate that the learned covariance structure captures known sources of label noise between semantically similar and co-occurring classes. Compared to standard neural network training and other baselines, we show significantly improved accuracy on Imagenet ILSVRC 2012 79.3% (+2.6%), Imagenet-21k 47.0% (+1.1%) and JFT 64.7% (+1.6%). We set a new state-of-the-art result on WebVision 1.0 with 76.6% top-1 accuracy. These datasets range from over 1M to over 300M training examples and from 1k classes to more than 21k classes. Our method is simple to use, and we provide an implementation that is a drop-in replacement for the final fully-connected layer in a deep classifier.
Lesion appearance is a crucial clue for medical providers to distinguish referable diabetic retinopathy (rDR) from non-referable DR. Most existing large-scale DR datasets contain only image-level labels rather than pixel-based annotations. This motivates us to develop algorithms to classify rDR and segment lesions via image-level labels. This paper leverages self-supervised equivariant learning and attention-based multi-instance learning (MIL) to tackle this problem. MIL is an effective strategy to differentiate positive and negative instances, helping us discard background regions (negative instances) while localizing lesion regions (positive ones). However, MIL only provides coarse lesion localization and cannot distinguish lesions located across adjacent patches. Conversely, a self-supervised equivariant attention mechanism (SEAM) generates a segmentation-level class activation map (CAM) that can guide patch extraction of lesions more accurately. Our work aims at integrating both methods to improve rDR classification accuracy. We conduct extensive validation experiments on the Eyepacs dataset, achieving an area under the receiver operating characteristic curve (AU ROC) of 0.958, outperforming current state-of-the-art algorithms.
We introduce equi-tuning, a novel fine-tuning method that transforms (potentially non-equivariant) pretrained models into group equivariant models while incurring minimum $L_2$ loss between the feature representations of the pretrained and the equivariant models. Large pretrained models can be equi-tuned for different groups to satisfy the needs of various downstream tasks. Equi-tuned models benefit from both group equivariance as an inductive bias and semantic priors from pretrained models. We provide applications of equi-tuning on three different tasks: image classification, compositional generalization in language, and fairness in natural language generation (NLG). We also provide a novel group-theoretic definition for fairness in NLG. The effectiveness of this definition is shown by testing it against a standard empirical method of fairness in NLG. We provide experimental results for equi-tuning using a variety of pretrained models: Alexnet, Resnet, VGG, and Densenet for image classification; RNNs, GRUs, and LSTMs for compositional generalization; and GPT2 for fairness in NLG. We test these models on benchmark datasets across all considered tasks to show the generality and effectiveness of the proposed method.
Transformation-robustness is an important feature for machine learning models that perform image classification. Many methods aim to bestow this property to models by the use of data augmentation strategies, while more formal guarantees are obtained via the use of equivariant models. We recognise that compositional, or part-whole structure is also an important aspect of images that has to be considered for building transformation-robust models. Thus, we propose a capsule network model that is, at once, equivariant and compositionality-aware. Equivariance of our capsule network model comes from the use of equivariant convolutions in a carefully-chosen novel architecture. The awareness of compositionality comes from the use of our proposed novel, iterative, graph-based routing algorithm, termed Iterative collaborative routing (ICR). ICR, the core of our contribution, weights the predictions made for capsules based on an iteratively averaged score of the degree-centralities of its nearest neighbours. Experiments on transformed image classification on FashionMNIST, CIFAR-10, and CIFAR-100 show that our model that uses ICR outperforms convolutional and capsule baselines to achieve state-of-the-art performance.
Object classification is plagued by the issue of session variation. Session variation describes any variation that makes one instance of an object look different to another, for instance due to pose or illumination variation. Recent work in the challenging task of face verification has shown that session variability modelling provides a mechanism to overcome some of these limitations. However, for computer vision purposes, it has only been applied in the limited setting of face verification. In this paper we propose a local region based inter-session variability (ISV) modelling approach, and apply it to challenging real-world data. We propose a region based session variability modelling approach so that local session variations can be modelled, termed Local ISV. We then demonstrate the efficacy of this technique on a challenging real-world fish image database which includes images taken underwater, providing significant real-world session variations. This Local ISV approach provides a relative performance improvement of, on average, 23% on the challenging MOBIO, Multi-PIE and SCface face databases. It also provides a relative performance improvement of 35% on our challenging fish image dataset
Foundation models have reshaped the landscape of Remote Sensing (RS) by enhancing various image interpretation tasks. Pretraining is an active research topic, encompassing supervised and self-supervised learning methods to initialize model weights effectively. However, transferring the pretrained models to downstream tasks may encounter task discrepancy due to their formulation of pretraining as image classification or object discrimination tasks. In this study, we explore the Multi-Task Pretraining (MTP) paradigm for RS foundation models to address this issue. Using a shared encoder and task-specific decoder architecture, we conduct multi-task supervised pretraining on the SAMRS dataset, encompassing semantic segmentation, instance segmentation, and rotated object detection. MTP supports both convolutional neural networks and vision transformer foundation models with over 300 million parameters. The pretrained models are finetuned on various RS downstream tasks, such as scene classification, horizontal and rotated object detection, semantic segmentation, and change detection. Extensive experiments across 14 datasets demonstrate the superiority of our models over existing ones of similar size and their competitive performance compared to larger state-of-the-art models, thus validating the effectiveness of MTP.
Incorporating pixel contextual information is critical for accurate segmentation. In this paper, we show that an effective way to incorporate contextual information is through a patch-based classifier. This patch classifier is trained to identify classes present within an image region, which facilitates the elimination of distractors and enhances the classification of small object segments. Specifically, we introduce Multi-scale Patch-based Multi-label Classifier (MPMC), a novel plug-in module designed for existing semi-supervised segmentation (SSS) frameworks. MPMC offers patch-level supervision, enabling the discrimination of pixel regions of different classes within a patch. Furthermore, MPMC learns an adaptive pseudo-label weight, using patch-level classification to alleviate the impact of the teacher's noisy pseudo-label supervision the student. This lightweight module can be integrated into any SSS framework, significantly enhancing their performance. We demonstrate the efficacy of our proposed MPMC by integrating it into four SSS methodologies and improving them across two natural image and one medical segmentation dataset, notably improving the segmentation results of the baselines across all the three datasets.
Large-scale image-text pre-trained models enable zero-shot classification and provide consistent accuracy across various data distributions. Nonetheless, optimizing these models in downstream tasks typically requires fine-tuning, which reduces generalization to out-of-distribution (OOD) data and demands extensive computational resources. We introduce Robust Adapter (R-Adapter), a novel method for fine-tuning zero-shot models to downstream tasks while simultaneously addressing both these issues. Our method integrates lightweight modules into the pre-trained model and employs novel self-ensemble techniques to boost OOD robustness and reduce storage expenses substantially. Furthermore, we propose MPM-NCE loss designed for fine-tuning on vision-language downstream tasks. It ensures precise alignment of multiple image-text pairs and discriminative feature learning. By extending the benchmark for robust fine-tuning beyond classification to include diverse tasks such as cross-modal retrieval and open vocabulary segmentation, we demonstrate the broad applicability of R-Adapter. Our extensive experiments demonstrate that R-Adapter achieves state-of-the-art performance across a diverse set of tasks, tuning only 13% of the parameters of the CLIP encoders.
Due to their effective performance, Convolutional Neural Network (CNN) and Vision Transformer (ViT) architectures have become the standard for solving computer vision tasks. Such architectures require large data sets and rely on convolution and self-attention operations. In 2021, MLP-Mixer emerged, an architecture that relies only on Multilayer Perceptron (MLP) and achieves extremely competitive results when compared to CNNs and ViTs. Despite its good performance in computer vision tasks, the MLP-Mixer architecture may not be suitable for refined feature extraction in images. Recently, the Kolmogorov-Arnold Network (KAN) was proposed as a promising alternative to MLP models. KANs promise to improve accuracy and interpretability when compared to MLPs. Therefore, the present work aims to design a new mixer-based architecture, called KAN-Mixers, using KANs as main layers and evaluate its performance, in terms of several performance metrics, in the image classification task. As main results obtained, the KAN-Mixers model was superior to the MLP, MLP-Mixer and KAN models in the Fashion-MNIST and CIFAR-10 datasets, with 0.9030 and 0.6980 of average accuracy, respectively.
This paper describes a novel approach to learning term-weighting schemes (TWSs) in the context of text classification. In text mining a TWS determines the way in which documents will be represented in a vector space model, before applying a classifier. Whereas acceptable performance has been obtained with standard TWSs (e.g., Boolean and term-frequency schemes), the definition of TWSs has been traditionally an art. Further, it is still a difficult task to determine what is the best TWS for a particular problem and it is not clear yet, whether better schemes, than those currently available, can be generated by combining known TWS. We propose in this article a genetic program that aims at learning effective TWSs that can improve the performance of current schemes in text classification. The genetic program learns how to combine a set of basic units to give rise to discriminative TWSs. We report an extensive experimental study comprising data sets from thematic and non-thematic text classification as well as from image classification. Our study shows the validity of the proposed method; in fact, we show that TWSs learned with the genetic program outperform traditional schemes and other TWSs proposed in recent works. Further, we show that TWSs learned from a specific domain can be effectively used for other tasks.
Remote sensing scene classification (RSSC) is a critical task with diverse applications in land use and resource management. While unimodal image-based approaches show promise, they often struggle with limitations such as high intra-class variance and inter-class similarity. Incorporating textual information can enhance classification by providing additional context and semantic understanding, but manual text annotation is labor-intensive and costly. In this work, we propose a novel RSSC framework that integrates text descriptions generated by large vision-language models (VLMs) as an auxiliary modality without incurring expensive manual annotation costs. To fully leverage the latent complementarities between visual and textual data, we propose a dual cross-attention-based network to fuse these modalities into a unified representation. Extensive experiments with both quantitative and qualitative evaluation across five RSSC datasets demonstrate that our framework consistently outperforms baseline models. We also verify the effectiveness of VLM-generated text descriptions compared to human-annotated descriptions. Additionally, we design a zero-shot classification scenario to show that the learned multimodal representation can be effectively utilized for unseen class classification. This research opens new opportunities for leveraging textual information in RSSC tasks and provides a promising multimodal fusion structure, offering insights and inspiration for future studies. Code is available at: https://github.com/CJR7/MultiAtt-RSSC
The integration of artificial intelligence (AI) with radiology marks a transformative era in medicine. Vision foundation models have been adopted to enhance radiologic imaging analysis. However, the distinct complexities of radiologic 2D and 3D radiologic data pose unique challenges that existing models, pre-trained on general non-medical images, fail to address adequately. To bridge this gap and capitalize on the diagnostic precision required in radiologic imaging, we introduce Radiologic Contrastive Language-Image Pre-training (RadCLIP): a cross-modal vision-language foundational model that harnesses Vision Language Pre-training (VLP) framework to improve radiologic image analysis. Building upon Contrastive Language-Image Pre-training (CLIP), RadCLIP incorporates a slice pooling mechanism tailored for volumetric image analysis and is pre-trained using a large and diverse dataset of radiologic image-text pairs. The RadCLIP was pre-trained to effectively align radiologic images with their corresponding text annotations, creating a robust vision backbone for radiologic images. Extensive experiments demonstrate RadCLIP's superior performance in both uni-modal radiologic image classification and cross-modal image-text matching, highlighting its significant promise for improving diagnostic accuracy and efficiency in clinical settings. Our Key contributions include curating a large dataset with diverse radiologic 2D/3D radiologic image-text pairs, a slice pooling adapter using an attention mechanism for integrating 2D images, and comprehensive evaluations of RadCLIP on various radiologic downstream tasks.
The problem of object localization has become one of the mainstream problems of vision. Most of the algorithms proposed involve the design for the model to be specifically for localizing objects. In this paper, we explore whether a pre-trained canonical ConvNet (without fine-tuning) trained purely for object classification on one dataset with global image level labels can be used to localize objects in images containing a single instance on a separate dataset while generalizing to novel classes. We propose a simple algorithm involving cropping and blackening out regions in the image space called Explicit Image Space based Search (EISS) for locating the most responsive regions in an image in the context of object localization. EISS brings to light the interesting phenomenon of a ConvNets responding more to features within objects as opposed to object level descriptors, as the classes in the training data get more correlated (visually/semantically similar).
Sparse decomposition has been widely used for different applications, such as source separation, image classification and image denoising. This paper presents a new algorithm for segmentation of an image into background and foreground text and graphics using sparse decomposition. First, the background is represented using a suitable smooth model, which is a linear combination of a few smoothly varying basis functions, and the foreground text and graphics are modeled as a sparse component overlaid on the smooth background. Then the background and foreground are separated using a sparse decomposition framework and imposing some prior information, which promote the smoothness of background, and the sparsity and connectivity of foreground pixels. This algorithm has been tested on a dataset of images extracted from HEVC standard test sequences for screen content coding, and is shown to outperform prior methods, including least absolute deviation fitting, k-means clustering based segmentation in DjVu, and shape primitive extraction and coding algorithm.
Local structures of shadow boundaries as well as complex interactions of image regions remain largely unexploited by previous shadow detection approaches. In this paper, we present a novel learning-based framework for shadow region recovery from a single image. We exploit the local structures of shadow edges by using a structured CNN learning framework. We show that using the structured label information in the classification can improve the local consistency of the results and avoid spurious labelling. We further propose and formulate a shadow/bright measure to model the complex interactions among image regions. The shadow and bright measures of each patch are computed from the shadow edges detected in the image. Using the global interaction constraints on patches, we formulate a least-square optimization problem for shadow recovery that can be solved efficiently. Our shadow recovery method achieves state-of-the-art results on the major shadow benchmark databases collected under various conditions.
In this paper, we propose an approach for learning binary hash codes for image retrieval. Canonical Correlation Analysis (CCA) is used to design two loss functions for training a neural network such that the correlation between the two views to CCA is maximized. The first loss, maximizes the correlation between the hash centers and learned hash codes. The second loss maximizes the correlation between the class labels and classification scores. A novel weighted mean and thresholding based hash center update scheme is proposed for adapting the hash centers in each epoch. The training loss reaches the theoretical lower bound of the proposed loss functions, showing that the correlation coefficients are maximized during training and substantiating the formation of an efficient feature space for image retrieval. The measured mean average precision shows that the proposed approach outperforms other state-of-the-art approaches in both single-labeled and multi-labeled image datasets.
Deep convolutional neural networks (CNNs) have achieved remarkable results in image processing tasks. However, their high expression ability risks overfitting. Consequently, data augmentation techniques have been proposed to prevent overfitting while enriching datasets. Recent CNN architectures with more parameters are rendering traditional data augmentation techniques insufficient. In this study, we propose a new data augmentation technique called random image cropping and patching (RICAP) which randomly crops four images and patches them to create a new training image. Moreover, RICAP mixes the class labels of the four images, resulting in an advantage similar to label smoothing. We evaluated RICAP with current state-of-the-art CNNs (e.g., the shake-shake regularization model) by comparison with competitive data augmentation techniques such as cutout and mixup. RICAP achieves a new state-of-the-art test error of $2.19\%$ on CIFAR-10. We also confirmed that deep CNNs with RICAP achieve better results on classification tasks using CIFAR-100 and ImageNet and an image-caption retrieval task using Microsoft COCO.
Medical vision--language models align radiographs with free-text reports, but radiology language distinguishes explicit findings, negation, uncertainty, and omission. Treating these report states as interchangeable during feature adaptation can change image predictions without supporting visual evidence. We introduce CARET, a parameter-efficient image-side adaptation method that retains pretrained image and text encoders and uses paired positive and negative clinical statements to define a low-dimensional semantic correction space. A bounded residual adjusts image features along these directions, while assertion-aware supervision distinguishes explicitly present and absent findings from uncertain or unmentioned ones. At inference, the adapter requires only an image; textual references remain fixed. In zero-shot classification, CARET achieves 67.80\% accuracy on CheXpert 5$\times$200 and 65.20\% on RSNA. These scores exceed the published CLEFT-LoRA results by 1.06 and 0.27 percentage points, respectively. Comparisons to literature baselines are informative but not controlled for differences in pretraining and evaluation protocols. Ablations across eight components identify the four-state distinction and assertion objective as major contributors, with task-dependent effects for nonmention preservation and pair alignment.