Recent advances in large vision-language models have revolutionized the image classification paradigm. Despite showing impressive zero-shot capabilities, a pre-defined set of categories, a.k.a. the vocabulary, is assumed at test time for composing the textual prompts. However, such assumption can be impractical when the semantic context is unknown and evolving. We thus formalize a novel task, termed as Vocabulary-free Image Classification (VIC), where we aim to assign to an input image a class that resides in an unconstrained language-induced semantic space, without the prerequisite of a known vocabulary. VIC is a challenging task as the semantic space is extremely large, containing millions of concepts, with hard-to-discriminate fine-grained categories. In this work, we first empirically verify that representing this semantic space by means of an external vision-language database is the most effective way to obtain semantically relevant content for classifying the image. We then propose Category Search from External Databases (CaSED), a method that exploits a pre-trained vision-language model and an external vision-language database to address VIC in a training-free manner. CaSED first extracts a set of candidate categories from captions retrieved from the database based on their semantic similarity to the image, and then assigns to the image the best matching candidate category according to the same vision-language model. Experiments on benchmark datasets validate that CaSED outperforms other complex vision-language frameworks, while being efficient with much fewer parameters, paving the way for future research in this direction.
Kernel-based approaches for sequence classification have been successfully applied to a variety of domains, including the text categorization, image classification, speech analysis, biological sequence analysis, time series and music classification, where they show some of the most accurate results. Typical kernel functions for sequences in these domains (e.g., bag-of-words, mismatch, or subsequence kernels) are restricted to {\em discrete univariate} (i.e. one-dimensional) string data, such as sequences of words in the text analysis, codeword sequences in the image analysis, or nucleotide or amino acid sequences in the DNA and protein sequence analysis. However, original sequence data are often of real-valued multivariate nature, i.e. are not univariate and discrete as required by typical $k$-mer based sequence kernel functions. In this work, we consider the problem of the {\em multivariate} sequence classification such as classification of multivariate music sequences, or multidimensional protein sequence representations. To this end, we extend {\em univariate} kernel functions typically used in sequence analysis and propose efficient {\em multivariate} similarity kernel method (MVDFQ-SK) based on (1) a direct feature quantization (DFQ) of each sequence dimension in the original {\em real-valued} multivariate sequences and (2) applying novel multivariate discrete kernel measures on these multivariate discrete DFQ sequence representations to more accurately capture similarity relationships among sequences and improve classification performance. Experiments using the proposed MVDFQ-SK kernel method show excellent classification performance on three challenging music classification tasks as well as protein sequence classification with significant 25-40% improvements over univariate kernel methods and existing state-of-the-art sequence classification methods.
Based on PixelHop and PixelHop++, which are recently developed using the successive subspace learning (SSL) framework, we propose an enhanced solution for object classification, called E-PixelHop, in this work. E-PixelHop consists of the following steps. First, to decouple the color channels for a color image, we apply principle component analysis and project RGB three color channels onto two principle subspaces which are processed separately for classification. Second, to address the importance of multi-scale features, we conduct pixel-level classification at each hop with various receptive fields. Third, to further improve pixel-level classification accuracy, we develop a supervised label smoothing (SLS) scheme to ensure prediction consistency. Forth, pixel-level decisions from each hop and from each color subspace are fused together for image-level decision. Fifth, to resolve confusing classes for further performance boosting, we formulate E-PixelHop as a two-stage pipeline. In the first stage, multi-class classification is performed to get a soft decision for each class, where the top 2 classes with the highest probabilities are called confusing classes. Then,we conduct a binary classification in the second stage. The main contributions lie in Steps 1, 3 and 5.We use the classification of the CIFAR-10 dataset as an example to demonstrate the effectiveness of the above-mentioned key components of E-PixelHop.
We conduct an in-depth exploration of different strategies for doing event detection in videos using convolutional neural networks (CNNs) trained for image classification. We study different ways of performing spatial and temporal pooling, feature normalization, choice of CNN layers as well as choice of classifiers. Making judicious choices along these dimensions led to a very significant increase in performance over more naive approaches that have been used till now. We evaluate our approach on the challenging TRECVID MED'14 dataset with two popular CNN architectures pretrained on ImageNet. On this MED'14 dataset, our methods, based entirely on image-trained CNN features, can outperform several state-of-the-art non-CNN models. Our proposed late fusion of CNN- and motion-based features can further increase the mean average precision (mAP) on MED'14 from 34.95% to 38.74%. The fusion approach achieves the state-of-the-art classification performance on the challenging UCF-101 dataset.
Self-supervised representation learning approaches have recently surpassed their supervised learning counterparts on downstream tasks like object detection and image classification. Somewhat mysteriously the recent gains in performance come from training instance classification models, treating each image and it's augmented versions as samples of a single class. In this work, we first present quantitative experiments to demystify these gains. We demonstrate that approaches like MOCO and PIRL learn occlusion-invariant representations. However, they fail to capture viewpoint and category instance invariance which are crucial components for object recognition. Second, we demonstrate that these approaches obtain further gains from access to a clean object-centric training dataset like Imagenet. Finally, we propose an approach to leverage unstructured videos to learn representations that possess higher viewpoint invariance. Our results show that the learned representations outperform MOCOv2 trained on the same data in terms of invariances encoded and the performance on downstream image classification and semantic segmentation tasks.
Active learning(AL) has recently gained popularity for deep learning(DL) models. This is due to efficient and informative sampling, especially when the learner requires large-scale labelled datasets. Commonly, the sampling and training happen in stages while more batches are added. One main bottleneck in this strategy is the narrow representation learned by the model that affects the overall AL selection. We present MoBYv2AL, a novel self-supervised active learning framework for image classification. Our contribution lies in lifting MoBY, one of the most successful self-supervised learning algorithms, to the AL pipeline. Thus, we add the downstream task-aware objective function and optimize it jointly with contrastive loss. Further, we derive a data-distribution selection function from labelling the new examples. Finally, we test and study our pipeline robustness and performance for image classification tasks. We successfully achieved state-of-the-art results when compared to recent AL methods. Code available: https://github.com/razvancaramalau/MoBYv2AL
The popularity of data augmentation techniques in machine learning has increased in recent years, as they enable the creation of new samples from existing datasets. Rotational augmentation, in particular, has shown great promise by revolving images and utilising them as additional data points for training. This research study introduces a new approach to enhance the performance of classification methods where the testing sets were generated employing transformations on every image from the original dataset. Subsequently, ensemble-based systems were implemented to determine the most reliable outcome in each subset acquired from the augmentation phase to get a final prediction for every original image. The findings of this study suggest that rotational augmentation techniques can significantly improve the accuracy of standard classification models; and the selection of a voting scheme can considerably impact the model's performance. Overall, the study found that using an ensemble-based voting system produced more accurate results than simple voting.
Dictionary learning algorithms have been successfully used in both reconstructive and discriminative tasks, where the input signal is represented by a linear combination of a few dictionary atoms. While these methods are usually developed under $\ell_1$ sparsity constrain (prior) in the input domain, recent studies have demonstrated the advantages of sparse representation using structured sparsity priors in the kernel domain. In this paper, we propose a supervised dictionary learning algorithm in the kernel domain for hyperspectral image classification. In the proposed formulation, the dictionary and classifier are obtained jointly for optimal classification performance. The supervised formulation is task-driven and provides learned features from the hyperspectral data that are well suited for the classification task. Moreover, the proposed algorithm uses a joint ($\ell_{12}$) sparsity prior to enforce collaboration among the neighboring pixels. The simulation results illustrate the efficiency of the proposed dictionary learning algorithm.
Deep learning based food image classification has enabled more accurate nutrition content analysis for image-based dietary assessment by predicting the types of food in eating occasion images. However, there are two major obstacles to apply food classification in real life applications. First, real life food images are usually heavy-tailed distributed, resulting in severe class-imbalance issue. Second, it is challenging to train a single-stage (i.e. end-to-end) framework under heavy-tailed data distribution, which cause the over-predictions towards head classes with rich instances and under-predictions towards tail classes with rare instance. In this work, we address both issues by introducing a novel single-stage heavy-tailed food classification framework. Our method is evaluated on two heavy-tailed food benchmark datasets, Food101-LT and VFN-LT, and achieves the best performance compared to existing work with over 5% improvements for top-1 accuracy.
Hyperspectral image (HSI) classification remains a challenging task due to the intricate spatial-spectral correlations. Existing transformer models excel in capturing long-range dependencies but often suffer from information redundancy and attention inefficiencies, limiting their ability to model fine-grained relationships crucial for HSI classification. To overcome these limitations, this work proposes MemFormer, a lightweight and memory-enhanced transformer. MemFormer introduces a memory-enhanced multi-head attention mechanism that iteratively refines a dynamic memory module, enhancing feature extraction while reducing redundancy across layers. Additionally, a dynamic memory enrichment strategy progressively captures complex spatial and spectral dependencies, leading to more expressive feature representations. To further improve structural consistency, we incorporate a spatial-spectral positional encoding (SSPE) tailored for HSI data, ensuring continuity without the computational burden of convolution-based approaches. Extensive experiments on benchmark datasets demonstrate that MemFormer achieves superior classification accuracy, outperforming state-of-the-art methods.
Joint Energy-based Model (JEM) of Grathwohl et al. shows that a standard softmax classifier can be reinterpreted as an energy-based model (EBM) for the joint distribution p(x,y); the resulting model can be optimized to improve calibration, robustness, and out-of-distribution detection, while generating samples rivaling the quality of recent GAN-based approaches. However, the softmax classifier that JEM exploits is inherently discriminative and its latent feature space is not well formulated as probabilistic distributions, which may hinder its potential for image generation and incur training instability. We hypothesize that generative classifiers, such as Linear Discriminant Analysis (LDA), might be more suitable for image generation since generative classifiers model the data generation process explicitly. This paper therefore investigates an LDA classifier for image classification and generation. In particular, the Max-Mahalanobis Classifier (MMC), a special case of LDA, fits our goal very well. We show that our Generative MMC (GMMC) can be trained discriminatively, generatively, or jointly for image classification and generation. Extensive experiments on multiple datasets show that GMMC achieves state-of-the-art discriminative and generative performances, while outperforming JEM in calibration, adversarial robustness, and out-of-distribution detection by a significant margin. Our source code is available at https://github.com/sndnyang/GMMC.
The bag-of-features (BoF) model for image classification has been thoroughly studied over the last decade. Different from the widely used BoF methods which modeled images with a pre-trained codebook, the alternative codebook free image modeling method, which we call Codebookless Model (CLM), attracted little attention. In this paper, we present an effective CLM that represents an image with a single Gaussian for classification. By embedding Gaussian manifold into a vector space, we show that the simple incorporation of our CLM into a linear classifier achieves very competitive accuracy compared with state-of-the-art BoF methods (e.g., Fisher Vector). Since our CLM lies in a high dimensional Riemannian manifold, we further propose a joint learning method of low-rank transformation with support vector machine (SVM) classifier on the Gaussian manifold, in order to reduce computational and storage cost. To study and alleviate the side effect of background clutter on our CLM, we also present a simple yet effective partial background removal method based on saliency detection. Experiments are extensively conducted on eight widely used databases to demonstrate the effectiveness and efficiency of our CLM method.
The unprecedented performance achieved by deep convolutional neural networks for image classification is linked primarily to their ability of capturing rich structural features at various layers within networks. Here we design a series of experiments, inspired by children's learning of the arithmetic addition of two integers, to showcase that such deep networks can go beyond the structural features to learn deeper knowledge. In our experiments, a set of images is constructed, each image containing an arithmetic addition $n+m$ in its central area, and several classification networks are then trained over a subset of images, using the sum as the label. Tests on the excluded images show that, as the image set gets larger, the networks have well learnt the law of arithmetic additions so as to build up their autonomous reasoning ability strongly. For instance, networks trained over a small percentage of images can classify a big majority of the remaining images correctly, and many arithmetic additions involving some integers that have never been seen during the training can also be solved correctly by the trained networks.
In the field of multi-source remote sensing image classification, remarkable progress has been made by using Convolutional Neural Network (CNN) and Transformer. Recently, Mamba-based methods built upon the State Space Model (SSM) have shown great potential for long-range dependency modeling with linear complexity, but they have rarely been explored for multi-source remote sensing image classification tasks. To address this issue, we propose the Multi-Scale Feature Fusion Mamba (MSFMamba) network, a novel framework designed for the joint classification of hyperspectral image (HSI) and Light Detection and Ranging (LiDAR)/Synthetic Aperture Radar (SAR) data. The MSFMamba network is composed of three key components: the Multi-Scale Spatial Mamba (MSpa-Mamba) block, the Spectral Mamba (Spe-Mamba) block, and the Fusion Mamba (Fus-Mamba) block. The MSpa-Mamba block employs a multi-scale strategy to reduce computational cost and alleviate feature redundancy in multiple scanning routes, ensuring efficient spatial feature modeling. The Spe-Mamba block focuses on spectral feature extraction, addressing the unique challenges of HSI data representation. Finally, the Fus-Mamba block bridges the heterogeneous gap between HSI and LiDAR/SAR data by extending the original Mamba architecture to accommodate dual inputs, enhancing cross-modal feature interactions and enabling seamless data fusion. Together, these components enable MSFMamba to effectively tackle the challenges of multi-source data classification, delivering improved performance with optimized computational efficiency. Comprehensive experiments on four real-world multi-source remote sensing datasets demonstrate the superiority of MSFMamba outperforms several state-of-the-art methods. The source codes of MSFMamba are publicly available at https://github.com/oucailab/MSFMamba.
In this paper, a novel statistical metric learning is developed for spectral-spatial classification of the hyperspectral image. First, the standard variance of the samples of each class in each batch is used to decrease the intra-class variance within each class. Then, the distances between the means of different classes are used to penalize the inter-class variance of the training samples. Finally, the standard variance between the means of different classes is added as an additional diversity term to repulse different classes from each other. Experiments have conducted over two real-world hyperspectral image datasets and the experimental results have shown the effectiveness of the proposed statistical metric learning.
This paper proposes a new methodology to automatically build semantic hierarchies suitable for image annotation and classification. The building of the hierarchy is based on a new measure of semantic similarity. The proposed measure incorporates several sources of information: visual, conceptual and contextual as we defined in this paper. The aim is to provide a measure that best represents image semantics. We then propose rules based on this measure, for the building of the final hierarchy, and which explicitly encode hierarchical relationships between different concepts. Therefore, the built hierarchy is used in a semantic hierarchical classification framework for image annotation. Our experiments and results show that the hierarchy built improves classification results. Ce papier propose une nouvelle methode pour la construction automatique de hierarchies semantiques adaptees a la classification et a l'annotation d'images. La construction de la hierarchie est basee sur une nouvelle mesure de similarite semantique qui integre plusieurs sources d'informations: visuelle, conceptuelle et contextuelle que nous definissons dans ce papier. L'objectif est de fournir une mesure qui est plus proche de la semantique des images. Nous proposons ensuite des regles, basees sur cette mesure, pour la construction de la hierarchie finale qui encode explicitement les relations hierarchiques entre les differents concepts. La hierarchie construite est ensuite utilisee dans un cadre de classification semantique hierarchique d'images en concepts visuels. Nos experiences et resultats montrent que la hierarchie construite permet d'ameliorer les resultats de la classification.
Through a simple multiple choice language prompt a VQA model can operate as a zero-shot image classifier, producing a classification label. Compared to typical image encoders, VQA models offer an advantage: VQA-produced image embeddings can be infused with the most relevant visual information through tailored language prompts. Nevertheless, for most tasks, zero-shot VQA performance is lacking, either because of unfamiliar category names, or dissimilar pre-training data and test data distributions. We propose a simple method to boost VQA performance for image classification using only a handful of labeled examples and a multiple-choice question. This few-shot method is training-free and maintains the dynamic and flexible advantages of the VQA model. Rather than relying on the final language output, our approach uses multiple-choice questions to extract prompt-specific latent representations, which are enriched with relevant visual information. These representations are combined to create a final overall image embedding, which is decoded via reference to latent class prototypes constructed from the few labeled examples. We demonstrate this method outperforms both pure visual encoders and zero-shot VQA baselines to achieve impressive performance on common few-shot tasks including MiniImageNet, Caltech-UCSD Birds, and CIFAR-100. Finally, we show our approach does particularly well in settings with numerous diverse visual attributes such as the fabric, article-style, texture, and view of different articles of clothing, where other few-shot approaches struggle, as we can tailor our image representations only on the semantic features of interest.
In this paper, we present a Neuron Abandoning Attention Flow (NAFlow) method to address the open problem of visually explaining the attention evolution dynamics inside CNNs when making their classification decisions. A novel cascading neuron abandoning back-propagation algorithm is designed to trace neurons in all layers of a CNN that involve in making its prediction to address the problem of significant interference from abandoned neurons. Firstly, a Neuron Abandoning Back-Propagation (NA-BP) module is proposed to generate Back-Propagated Feature Maps (BPFM) by using the inverse function of the intermediate layers of CNN models, on which the neurons not used for decision-making are abandoned. Meanwhile, the cascading NA-BP modules calculate the tensors of importance coefficients which are linearly combined with the tensors of BPFMs to form the NAFlow. Secondly, to be able to visualize attention flow for similarity metric-based CNN models, a new channel contribution weights module is proposed to calculate the importance coefficients via Jacobian Matrix. The effectiveness of the proposed NAFlow is validated on nine widely-used CNN models for various tasks of general image classification, contrastive learning classification, few-shot image classification, and image retrieval.
In this paper, we propose a convolutional neural network with mapping layers (MCNN) for hyperspectral image (HSI) classification. The proposed mapping layers map the input patch into a low dimensional subspace by multilinear algebra. We use our mapping layers to reduce the spectral and spatial redundancy and maintain most energy of the input. The feature extracted by our mapping layers can also reduce the number of following convolutional layers for feature extraction. Our MCNN architecture avoids the declining accuracy with increasing layers phenomenon of deep learning models for HSI classification and also saves the training time for its effective mapping layers. Furthermore, we impose the 3-D convolutional kernel on convolutional layer to extract the spectral-spatial features for HSI. We tested our MCNN on three datasets of Indian Pines, University of Pavia and Salinas, and we achieved the classification accuracy of 98.3%, 99.5% and 99.3%, respectively. Experimental results demonstrate that the proposed MCNN can significantly improve the classification accuracy and save much time consumption.
Whole slide image (WSI) classification is an essential task in computational pathology. Despite the recent advances in multiple instance learning (MIL) for WSI classification, accurate classification of WSIs remains challenging due to the extreme imbalance between the positive and negative instances in bags, and the complicated pre-processing to fuse multi-scale information of WSI. To this end, we propose a novel multi-scale prototypical Transformer (MSPT) for WSI classification, which includes a prototypical Transformer (PT) module and a multi-scale feature fusion module (MFFM). The PT is developed to reduce redundant instances in bags by integrating prototypical learning into the Transformer architecture. It substitutes all instances with cluster prototypes, which are then re-calibrated through the self-attention mechanism of the Trans-former. Thereafter, an MFFM is proposed to fuse the clustered prototypes of different scales, which employs MLP-Mixer to enhance the information communication between prototypes. The experimental results on two public WSI datasets demonstrate that the proposed MSPT outperforms all the compared algorithms, suggesting its potential applications.