In real-world scenarios, image recognition tasks, such as semantic segmentation and object detection, often pose greater challenges due to the lack of information available within low-resolution (LR) content. Image super-resolution (SR) is one of the promising solutions for addressing the challenges. However, due to the ill-posed property of SR, it is challenging for typical SR methods to restore task-relevant high-frequency contents, which may dilute the advantage of utilizing the SR method. Therefore, in this paper, we propose Super-Resolution for Image Recognition (SR4IR) that effectively guides the generation of SR images beneficial to achieving satisfactory image recognition performance when processing LR images. The critical component of our SR4IR is the task-driven perceptual (TDP) loss that enables the SR network to acquire task-specific knowledge from a network tailored for a specific task. Moreover, we propose a cross-quality patch mix and an alternate training framework that significantly enhances the efficacy of the TDP loss by addressing potential problems when employing the TDP loss. Through extensive experiments, we demonstrate that our SR4IR achieves outstanding task performance by generating SR images useful for a specific image recognition task, including semantic segmentation, object detection, and image classification. The implementation code is available at https://github.com/JaehaKim97/SR4IR.
Hyperspectral image (HSI) classification techniques have been intensively studied and a variety of models have been developed. However, these HSI classification models are confined to pocket models and unrealistic ways of dataset partitioning. The former limits the generalization performance of the model and the latter is partitioned leading to inflated model evaluation metrics, which results in plummeting model performance in the real world. Therefore, we propose a universal knowledge embedded contrastive learning framework (KnowCL) for supervised, unsupervised, and semisupervised HSI classification, which largely closes the gap between HSI classification models between pocket models and standard vision backbones. We present a new HSI processing pipeline in conjunction with a range of data transformation and augmentation techniques that provide diverse data representations and realistic data partitioning. The proposed framework based on this pipeline is compatible with all kinds of backbones and can fully exploit labeled and unlabeled samples with the expected training time. Furthermore, we design a new loss function, which can adaptively fuse the supervised loss and unsupervised loss, enhancing the learning performance. This proposed new classification paradigm shows great potential in exploring for HSI classification technology. The code can be accessed at \url{https://github.com/quanweiliu/KnowCL}.
With the increasing number and enhanced capabilities of IoT devices in smart buildings, these devices are evolving beyond basic data collection and control to actively participate in deep learning tasks. Federated Learning (FL), as a decentralized learning paradigm, is well-suited for such scenarios. However, the limited computational and communication resources of IoT devices present significant challenges. While existing research has extensively explored efficiency improvements in Horizontal FL, these techniques cannot be directly applied to Vertical FL due to fundamental differences in data partitioning and model structure. To address this gap, we propose a Lightweight Vertical Federated Learning (LVFL) framework that jointly optimizes computational and communication efficiency. Our approach introduces two distinct lightweighting strategies: one for reducing the complexity of the feature model to improve local computation, and another for compressing feature embeddings to reduce communication overhead. Furthermore, we derive a convergence bound for the proposed LVFL algorithm that explicitly incorporates both computation and communication lightweighting ratios. Experimental results on an image classification task demonstrate that LVFL effectively mitigates resource demands while maintaining competitive learning performance.
Post-training quantization (PTQ) is an efficient model compression technique that quantizes a pretrained full-precision model using only a small calibration set of unlabeled samples without retraining. PTQ methods for convolutional neural networks (CNNs) provide quantization results comparable to full-precision counterparts. Directly applying them to vision transformers (ViTs), however, incurs severe performance degradation, mainly due to the differences in architectures between CNNs and ViTs. In particular, the distribution of activations for each channel vary drastically according to input instances, making PTQ methods for CNNs inappropriate for ViTs. To address this, we introduce instance-aware group quantization for ViTs (IGQ-ViT). To this end, we propose to split the channels of activation maps into multiple groups dynamically for each input instance, such that activations within each group share similar statistical properties. We also extend our scheme to quantize softmax attentions across tokens. In addition, the number of groups for each layer is adjusted to minimize the discrepancies between predictions from quantized and full-precision models, under a bit-operation (BOP) constraint. We show extensive experimental results on image classification, object detection, and instance segmentation, with various transformer architectures, demonstrating the effectiveness of our approach.
Large-scale contrastive vision-language pre-trained models provide the zero-shot model achieving competitive performance across a range of image classification tasks without requiring training on downstream data. Recent works have confirmed that while additional fine-tuning of the zero-shot model on the reference data results in enhanced downstream performance, it compromises the model's robustness against distribution shifts. Our investigation begins by examining the conditions required to achieve the goals of robust fine-tuning, employing descriptions based on feature distortion theory and joint energy-based models. Subsequently, we propose a novel robust fine-tuning algorithm, Lipsum-FT, that effectively utilizes the language modeling aspect of the vision-language pre-trained models. Extensive experiments conducted on distribution shift scenarios in DomainNet and ImageNet confirm the superiority of our proposed Lipsum-FT approach over existing robust fine-tuning methods.
CAM-based methods are widely-used post-hoc interpretability method that produce a saliency map to explain the decision of an image classification model. The saliency map highlights the important areas of the image relevant to the prediction. In this paper, we show that most of these methods can incorrectly attribute an important score to parts of the image that the model cannot see. We show that this phenomenon occurs both theoretically and experimentally. On the theory side, we analyze the behavior of GradCAM on a simple masked CNN model at initialization. Experimentally, we train a VGG-like model constrained to not use the lower part of the image and nevertheless observe positive scores in the unseen part of the image. This behavior is evaluated quantitatively on two new datasets. We believe that this is problematic, potentially leading to mis-interpretation of the model's behavior.
Today, skin cancer is considered as one of the most dangerous and common cancers in the world which demands special attention. Skin cancer may be developed in different types; including melanoma, actinic keratosis, basal cell carcinoma, squamous cell carcinoma, and Merkel cell carcinoma. Among them, melanoma is more unpredictable. Melanoma cancer can be diagnosed at early stages increasing the possibility of disease treatment. Automatic classification of skin lesions is a challenging task due to diverse forms and grades of the disease, demanding the requirement of novel methods implementation. Deep convolution neural networks (CNN) have shown an excellent potential for data and image classification. In this article, we inspect skin lesion classification problem using CNN techniques. Remarkably, we present that prominent classification accuracy of lesion detection can be obtained by proper designing and applying of transfer learning framework on pre-trained neural networks, without any requirement for data enlargement procedures i.e. merging VGG16 and VGG19 architectures pre-trained by a generic dataset with modified AlexNet network, and then, fine-tuned by a subject-specific dataset containing dermatology images. The convolution neural network was trained using 2541 images and, in particular, dropout was used to prevent the network from overfitting. Finally, the validity of the model was checked by applying the K-fold cross validation method. The proposed model increased classification accuracy by 3% (from 94.2% to 98.18%) in comparison with other methods.
Visual scenes are naturally organized in a hierarchy, where a coarse semantic is recursively comprised of several fine details. Exploring such a visual hierarchy is crucial to recognize the complex relations of visual elements, leading to a comprehensive scene understanding. In this paper, we propose a Visual Hierarchy Mapper (Hi-Mapper), a novel approach for enhancing the structured understanding of the pre-trained Deep Neural Networks (DNNs). Hi-Mapper investigates the hierarchical organization of the visual scene by 1) pre-defining a hierarchy tree through the encapsulation of probability densities; and 2) learning the hierarchical relations in hyperbolic space with a novel hierarchical contrastive loss. The pre-defined hierarchy tree recursively interacts with the visual features of the pre-trained DNNs through hierarchy decomposition and encoding procedures, thereby effectively identifying the visual hierarchy and enhancing the recognition of an entire scene. Extensive experiments demonstrate that Hi-Mapper significantly enhances the representation capability of DNNs, leading to an improved performance on various tasks, including image classification and dense prediction tasks.
Biomedical image analysis is of paramount importance for the advancement of healthcare and medical research. Although conventional convolutional neural networks (CNNs) are frequently employed in this domain, facing limitations in capturing intricate spatial and temporal relationships at the pixel level due to their reliance on fixed-sized windows and immutable filter weights post-training. These constraints impede their ability to adapt to input fluctuations and comprehend extensive long-range contextual information. To overcome these challenges, a novel architecture based on self-attention mechanisms as an alternative to conventional CNNs.The proposed model utilizes attention-based mechanisms to surpass the limitations of CNNs. The key component of our strategy is the combination of non-overlapping (vanilla patching) and novel overlapped Shifted Patching Techniques (S.P.T.s), which enhances the model's capacity to capture local context and improves generalization. Additionally, we introduce the Lancoz5 interpolation technique, which adapts variable image sizes to higher resolutions, facilitating better analysis of high-resolution biomedical images. Our methods address critical challenges faced by attention-based vision models, including inductive bias, weight sharing, receptive field limitations, and efficient data handling. Experimental evidence shows the effectiveness of proposed model in generalizing to various biomedical imaging tasks. The attention-based model, combined with advanced data augmentation methodologies, exhibits robust modeling capabilities and superior performance compared to existing approaches. The integration of S.P.T.s significantly enhances the model's ability to capture local context, while the Lancoz5 interpolation technique ensures efficient handling of high-resolution images.
In this study, we present an advanced convolutional neural network (CNN) architecture for ship classification based on optical satellite imagery, which significantly enhances performance through the integration of a convolutional block attention module (CBAM) and additional architectural innovations. Building upon the foundational ResNet50 model, we first incorporated a standard CBAM to direct the model's focus toward more informative features, achieving an accuracy of 87% compared to 85% of the baseline ResNet50. Further augmentations involved multiscale feature integration, depthwise separable convolutions, and dilated convolutions, culminating in an enhanced ResNet model with improved CBAM. This model demonstrated a remarkable accuracy of 95%, with precision, recall, and F1 scores all witnessing substantial improvements across various ship classes. In particular, the bulk carrier and oil tanker classes exhibited nearly perfect precision and recall rates, underscoring the enhanced capability of the model to accurately identify and classify ships. Attention heatmap analyses further validated the efficacy of the improved model, revealing more focused attention on relevant ship features regardless of background complexities. These findings underscore the potential of integrating attention mechanisms and architectural innovations into CNNs for high-resolution satellite imagery classification. This study navigates through the class imbalance and computational costs and proposes future directions for scalability and adaptability in new or rare ship-type recognition. This study lays the groundwork for applying advanced deep learning techniques in remote sensing, offering insights into scalable and efficient satellite image classification.
We introduce ImageNot, a dataset designed to match the scale of ImageNet while differing drastically in other aspects. We show that key model architectures developed for ImageNet over the years rank identically when trained and evaluated on ImageNot to how they rank on ImageNet. This is true when training models from scratch or fine-tuning them. Moreover, the relative improvements of each model over earlier models strongly correlate in both datasets. We further give evidence that ImageNot has a similar utility as ImageNet for transfer learning purposes. Our work demonstrates a surprising degree of external validity in the relative performance of image classification models. This stands in contrast with absolute accuracy numbers that typically drop sharply even under small changes to a dataset.
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.
The ability to detect unfamiliar or unexpected images is essential for safe deployment of computer vision systems. In the context of classification, the task of detecting images outside of a model's training domain is known as out-of-distribution (OOD) detection. While there has been a growing research interest in developing post-hoc OOD detection methods, there has been comparably little discussion around how these methods perform when the underlying classifier is not trained on a clean, carefully curated dataset. In this work, we take a closer look at 20 state-of-the-art OOD detection methods in the (more realistic) scenario where the labels used to train the underlying classifier are unreliable (e.g. crowd-sourced or web-scraped labels). Extensive experiments across different datasets, noise types & levels, architectures and checkpointing strategies provide insights into the effect of class label noise on OOD detection, and show that poor separation between incorrectly classified ID samples vs. OOD samples is an overlooked yet important limitation of existing methods. Code: https://github.com/glhr/ood-labelnoise
The robust generalization of models to rare, in-distribution (ID) samples drawn from the long tail of the training distribution and to out-of-training-distribution (OOD) samples is one of the major challenges of current deep learning methods. For image classification, this manifests in the existence of adversarial attacks, the performance drops on distorted images, and a lack of generalization to concepts such as sketches. The current understanding of generalization in neural networks is very limited, but some biases that differentiate models from human vision have been identified and might be causing these limitations. Consequently, several attempts with varying success have been made to reduce these biases during training to improve generalization. We take a step back and sanity-check these attempts. Fixing the architecture to the well-established ResNet-50, we perform a large-scale study on 48 ImageNet models obtained via different training methods to understand how and if these biases - including shape bias, spectral biases, and critical bands - interact with generalization. Our extensive study results reveal that contrary to previous findings, these biases are insufficient to accurately predict the generalization of a model holistically. We provide access to all checkpoints and evaluation code at https://github.com/paulgavrikov/biases_vs_generalization
Existing vision-language models (VLMs) treat text descriptions as a unit, confusing individual concepts in a prompt and impairing visual semantic matching and reasoning. An important aspect of reasoning in logic and language is negations. This paper highlights the limitations of popular VLMs such as CLIP, at understanding the implications of negations, i.e., the effect of the word "not" in a given prompt. To enable evaluation of VLMs on fluent prompts with negations, we present CC-Neg, a dataset containing 228,246 images, true captions and their corresponding negated captions. Using CC-Neg along with modifications to the contrastive loss of CLIP, our proposed CoN-CLIP framework, has an improved understanding of negations. This training paradigm improves CoN-CLIP's ability to encode semantics reliably, resulting in 3.85% average gain in top-1 accuracy for zero-shot image classification across 8 datasets. Further, CoN-CLIP outperforms CLIP on challenging compositionality benchmarks such as SugarCREPE by 4.4%, showcasing emergent compositional understanding of objects, relations, and attributes in text. Overall, our work addresses a crucial limitation of VLMs by introducing a dataset and framework that strengthens semantic associations between images and text, demonstrating improved large-scale foundation models with significantly reduced computational cost, promoting efficiency and accessibility.
Aiming at the metro video surveillance system has not been able to effectively solve the metro crowd density estimation problem, a Metro Crowd density estimation Network (called MCNet) is proposed to automatically classify crowd density level of passengers. Firstly, an Integrating Multi-scale Attention (IMA) module is proposed to enhance the ability of the plain classifiers to extract semantic crowd texture features to accommodate to the characteristics of the crowd texture feature. The innovation of the IMA module is to fuse the dilation convolution, multiscale feature extraction and attention mechanism to obtain multi-scale crowd feature activation from a larger receptive field with lower computational cost, and to strengthen the crowds activation state of convolutional features in top layers. Secondly, a novel lightweight crowd texture feature extraction network is proposed, which can directly process video frames and automatically extract texture features for crowd density estimation, while its faster image processing speed and fewer network parameters make it flexible to be deployed on embedded platforms with limited hardware resources. Finally, this paper integrates IMA module and the lightweight crowd texture feature extraction network to construct the MCNet, and validate the feasibility of this network on image classification dataset: Cifar10 and four crowd density datasets: PETS2009, Mall, QUT and SH_METRO to validate the MCNet whether can be a suitable solution for crowd density estimation in metro video surveillance where there are image processing challenges such as high density, high occlusion, perspective distortion and limited hardware resources.
Our paper addresses the problem of models struggling to learn diverse features, due to either forgetting previously learned features or failing to learn new ones. To overcome this problem, we introduce Diverse Feature Learning (DFL), a method that combines an important feature preservation algorithm with a new feature learning algorithm. Specifically, for preserving important features, we utilize self-distillation in ensemble models by selecting the meaningful model weights observed during training. For learning new features, we employ reset that involves periodically re-initializing part of the model. As a result, through experiments with various models on the image classification, we have identified the potential for synergistic effects between self-distillation and reset.
Recent advances in deep learning have mainly relied on Transformers due to their data dependency and ability to learn at scale. The attention module in these architectures, however, exhibits quadratic time and space in input size, limiting their scalability for long-sequence modeling. Despite recent attempts to design efficient and effective architecture backbone for multi-dimensional data, such as images and multivariate time series, existing models are either data independent, or fail to allow inter- and intra-dimension communication. Recently, State Space Models (SSMs), and more specifically Selective State Space Models, with efficient hardware-aware implementation, have shown promising potential for long sequence modeling. Motivated by the success of SSMs, we present MambaMixer, a new architecture with data-dependent weights that uses a dual selection mechanism across tokens and channels, called Selective Token and Channel Mixer. MambaMixer connects selective mixers using a weighted averaging mechanism, allowing layers to have direct access to early features. As a proof of concept, we design Vision MambaMixer (ViM2) and Time Series MambaMixer (TSM2) architectures based on the MambaMixer block and explore their performance in various vision and time series forecasting tasks. Our results underline the importance of selective mixing across both tokens and channels. In ImageNet classification, object detection, and semantic segmentation tasks, ViM2 achieves competitive performance with well-established vision models and outperforms SSM-based vision models. In time series forecasting, TSM2 achieves outstanding performance compared to state-of-the-art methods while demonstrating significantly improved computational cost. These results show that while Transformers, cross-channel attention, and MLPs are sufficient for good performance in time series forecasting, neither is necessary.
Polarimetric synthetic aperture radar (PolSAR) image interpretation is widely used in various fields. Recently, deep learning has made significant progress in PolSAR image classification. Supervised learning (SL) requires a large amount of labeled PolSAR data with high quality to achieve better performance, however, manually labeled data is insufficient. This causes the SL to fail into overfitting and degrades its generalization performance. Furthermore, the scattering confusion problem is also a significant challenge that attracts more attention. To solve these problems, this article proposes a Heterogeneous Network based Contrastive Learning method(HCLNet). It aims to learn high-level representation from unlabeled PolSAR data for few-shot classification according to multi-features and superpixels. Beyond the conventional CL, HCLNet introduces the heterogeneous architecture for the first time to utilize heterogeneous PolSAR features better. And it develops two easy-to-use plugins to narrow the domain gap between optics and PolSAR, including feature filter and superpixel-based instance discrimination, which the former is used to enhance the complementarity of multi-features, and the latter is used to increase the diversity of negative samples. Experiments demonstrate the superiority of HCLNet on three widely used PolSAR benchmark datasets compared with state-of-the-art methods. Ablation studies also verify the importance of each component. Besides, this work has implications for how to efficiently utilize the multi-features of PolSAR data to learn better high-level representation in CL and how to construct networks suitable for PolSAR data better.
Attention based Large Language Models (LLMs) are the state-of-the-art in natural language processing (NLP). The two most common architectures are encoders such as BERT, and decoders like the GPT models. Despite the success of encoder models, on which we focus in this work, they also bear several risks, including issues with bias or their susceptibility for adversarial attacks, signifying the necessity for explainable AI to detect such issues. While there does exist various local explainability methods focusing on the prediction of single inputs, global methods based on dimensionality reduction for classification inspection, which have emerged in other domains and that go further than just using t-SNE in the embedding space, are not widely spread in NLP. To reduce this gap, we investigate the application of DeepView, a method for visualizing a part of the decision function together with a data set in two dimensions, to the NLP domain. While in previous work, DeepView has been used to inspect deep image classification models, we demonstrate how to apply it to BERT-based NLP classifiers and investigate its usability in this domain, including settings with adversarially perturbed input samples and pre-trained, fine-tuned, and multi-task models.