In the field of medical microscopic image classification (MIC), CNN-based and Transformer-based models have been extensively studied. However, CNNs struggle with modeling long-range dependencies, limiting their ability to fully utilize semantic information in images. Conversely, Transformers are hampered by the complexity of quadratic computations. To address these challenges, we propose a model based on the Mamba architecture: Microscopic-Mamba. Specifically, we designed the Partially Selected Feed-Forward Network (PSFFN) to replace the last linear layer of the Visual State Space Module (VSSM), enhancing Mamba's local feature extraction capabilities. Additionally, we introduced the Modulation Interaction Feature Aggregation (MIFA) module to effectively modulate and dynamically aggregate global and local features. We also incorporated a parallel VSSM mechanism to improve inter-channel information interaction while reducing the number of parameters. Extensive experiments have demonstrated that our method achieves state-of-the-art performance on five public datasets. Code is available at https://github.com/zs1314/Microscopic-Mamba
We propose Vision Token Turing Machines (ViTTM), an efficient, low-latency, memory-augmented Vision Transformer (ViT). Our approach builds on Neural Turing Machines and Token Turing Machines, which were applied to NLP and sequential visual understanding tasks. ViTTMs are designed for non-sequential computer vision tasks such as image classification and segmentation. Our model creates two sets of tokens: process tokens and memory tokens; process tokens pass through encoder blocks and read-write from memory tokens at each encoder block in the network, allowing them to store and retrieve information from memory. By ensuring that there are fewer process tokens than memory tokens, we are able to reduce the inference time of the network while maintaining its accuracy. On ImageNet-1K, the state-of-the-art ViT-B has median latency of 529.5ms and 81.0% accuracy, while our ViTTM-B is 56% faster (234.1ms), with 2.4 times fewer FLOPs, with an accuracy of 82.9%. On ADE20K semantic segmentation, ViT-B achieves 45.65mIoU at 13.8 frame-per-second (FPS) whereas our ViTTM-B model acheives a 45.17 mIoU with 26.8 FPS (+94%).
In the context of few-shot classification, the goal is to train a classifier using a limited number of samples while maintaining satisfactory performance. However, traditional metric-based methods exhibit certain limitations in achieving this objective. These methods typically rely on a single distance value between the query feature and support feature, thereby overlooking the contribution of shallow features. To overcome this challenge, we propose a novel approach in this paper. Our approach involves utilizing a multi-output embedding network that maps samples into distinct feature spaces. The proposed method extracts feature vectors at different stages, enabling the model to capture both global and abstract features. By utilizing these diverse feature spaces, our model enhances its performance. Moreover, employing a self-attention mechanism improves the refinement of features at each stage, leading to even more robust representations and improved overall performance. Furthermore, assigning learnable weights to each stage significantly improved performance and results. We conducted comprehensive evaluations on the MiniImageNet and FC100 datasets, specifically in the 5-way 1-shot and 5-way 5-shot scenarios. Additionally, we performed cross-domain tasks across eight benchmark datasets, achieving high accuracy in the testing domains. These evaluations demonstrate the efficacy of our proposed method in comparison to state-of-the-art approaches. https://github.com/FatemehAskari/MSENet
Data augmentation (DA) has been widely used to improve the generalization of deep neural networks. While existing DA methods have proven effective, they often rely on augmentation operations with random magnitudes to each sample. However, this approach can inadvertently introduce noise, induce distribution shifts, and increase the risk of overfitting. In this paper, we propose EntAugment, a tuning-free and adaptive DA framework. Unlike previous work, EntAugment dynamically assesses and adjusts the augmentation magnitudes for each sample during training, leveraging insights into both the inherent complexities of training samples and the evolving status of deep models. Specifically, in EntAugment, the magnitudes are determined by the information entropy derived from the probability distribution obtained by applying the softmax function to the model's output. In addition, to further enhance the efficacy of EntAugment, we introduce a novel entropy regularization term, EntLoss, which complements the EntAugment approach. Theoretical analysis further demonstrates that EntLoss, compared to traditional cross-entropy loss, achieves closer alignment between the model distributions and underlying dataset distributions. Moreover, EntAugment and EntLoss can be utilized separately or jointly. We conduct extensive experiments across multiple image classification tasks and network architectures with thorough comparisons of existing DA methods. Importantly, the proposed methods outperform others without introducing any auxiliary models or noticeable extra computational costs, highlighting both effectiveness and efficiency. Code is available at https://github.com/Jackbrocp/EntAugment.
Data attribution aims to quantify the contribution of individual training data points to the outputs of an AI model, which has been used to measure the value of training data and compensate data providers. Given the impact on financial decisions and compensation mechanisms, a critical question arises concerning the adversarial robustness of data attribution methods. However, there has been little to no systematic research addressing this issue. In this work, we aim to bridge this gap by detailing a threat model with clear assumptions about the adversary's goal and capabilities and proposing principled adversarial attack methods on data attribution. We present two methods, Shadow Attack and Outlier Attack, which generate manipulated datasets to inflate the compensation adversarially. The Shadow Attack leverages knowledge about the data distribution in the AI applications, and derives adversarial perturbations through "shadow training", a technique commonly used in membership inference attacks. In contrast, the Outlier Attack does not assume any knowledge about the data distribution and relies solely on black-box queries to the target model's predictions. It exploits an inductive bias present in many data attribution methods - outlier data points are more likely to be influential - and employs adversarial examples to generate manipulated datasets. Empirically, in image classification and text generation tasks, the Shadow Attack can inflate the data-attribution-based compensation by at least 200%, while the Outlier Attack achieves compensation inflation ranging from 185% to as much as 643%. Our implementation is ready at https://github.com/TRAIS-Lab/adversarial-attack-data-attribution.
Deep learning models have achieved great success in automating skin lesion diagnosis. However, the ethnic disparity in these models' predictions needs to be addressed before deploying them. We introduce a novel approach, PatchAlign, to enhance skin condition image classification accuracy and fairness by aligning with clinical text representations of skin conditions. PatchAlign uses Graph Optimal Transport (GOT) Loss as a regularizer to perform cross-domain alignment. The representations obtained are robust and generalize well across skin tones, even with limited training samples. To reduce the effect of noise and artifacts in clinical dermatology images, we propose a learnable Masked Graph Optimal Transport for cross-domain alignment that further improves fairness metrics. We compare our model to the state-of-the-art FairDisCo on two skin lesion datasets with different skin types: Fitzpatrick17k and Diverse Dermatology Images (DDI). PatchAlign enhances the accuracy of skin condition image classification by 2.8% (in-domain) and 6.2% (out-domain) on Fitzpatrick17k, and 4.2% (in-domain) on DDI compared to FairDisCo. Additionally, it consistently improves the fairness of true positive rates across skin tones. The source code for the implementation is available at the following GitHub repository: https://github.com/aayushmanace/PatchAlign24, enabling easy reproduction and further experimentation.
As Deep Neural Networks have achieved thrilling breakthroughs in the past decade, data augmentations have garnered increasing attention as regularization techniques when massive labeled data are unavailable. Among existing augmentations, Mixup and relevant data-mixing methods that convexly combine selected samples and the corresponding labels are widely adopted because they yield high performances by generating data-dependent virtual data while easily migrating to various domains. This survey presents a comprehensive review of foundational mixup methods and their applications. We first elaborate on the training pipeline with mixup augmentations as a unified framework containing modules. A reformulated framework could contain various mixup methods and give intuitive operational procedures. Then, we systematically investigate the applications of mixup augmentations on vision downstream tasks, various data modalities, and some analysis \& theorems of mixup. Meanwhile, we conclude the current status and limitations of mixup research and point out further work for effective and efficient mixup augmentations. This survey can provide researchers with the current state of the art in mixup methods and provide some insights and guidance roles in the mixup arena. An online project with this survey is available at \url{https://github.com/Westlake-AI/Awesome-Mixup}.
Activation function has a significant impact on the dynamics, convergence, and performance of deep neural networks. The search for a consistent and high-performing activation function has always been a pursuit during deep learning model development. Existing state-of-the-art activation functions are manually designed with human expertise except for Swish. Swish was developed using a reinforcement learning-based search strategy. In this study, we propose an evolutionary approach for optimizing activation functions specifically for image classification tasks, aiming to discover functions that outperform current state-of-the-art options. Through this optimization framework, we obtain a series of high-performing activation functions denoted as Exponential Error Linear Unit (EELU). The developed activation functions are evaluated for image classification tasks from two perspectives: (1) five state-of-the-art neural network architectures, such as ResNet50, AlexNet, VGG16, MobileNet, and Compact Convolutional Transformer which cover computationally heavy to light neural networks, and (2) eight standard datasets, including CIFAR10, Imagenette, MNIST, Fashion MNIST, Beans, Colorectal Histology, CottonWeedID15, and TinyImageNet which cover from typical machine vision benchmark, agricultural image applications to medical image applications. Finally, we statistically investigate the generalization of the resultant activation functions developed through the optimization scheme. With a Friedman test, we conclude that the optimization scheme is able to generate activation functions that outperform the existing standard ones in 92.8% cases among 28 different cases studied, and $-x\cdot erf(e^{-x})$ is found to be the best activation function for image classification generated by the optimization scheme.
Plant diseases pose significant threats to agriculture. It necessitates proper diagnosis and effective treatment to safeguard crop yields. To automate the diagnosis process, image segmentation is usually adopted for precisely identifying diseased regions, thereby advancing precision agriculture. Developing robust image segmentation models for plant diseases demands high-quality annotations across numerous images. However, existing plant disease datasets typically lack segmentation labels and are often confined to controlled laboratory settings, which do not adequately reflect the complexity of natural environments. Motivated by this fact, we established PlantSeg, a large-scale segmentation dataset for plant diseases. PlantSeg distinguishes itself from existing datasets in three key aspects. (1) Annotation type: Unlike the majority of existing datasets that only contain class labels or bounding boxes, each image in PlantSeg includes detailed and high-quality segmentation masks, associated with plant types and disease names. (2) Image source: Unlike typical datasets that contain images from laboratory settings, PlantSeg primarily comprises in-the-wild plant disease images. This choice enhances the practical applicability, as the trained models can be applied for integrated disease management. (3) Scale: PlantSeg is extensive, featuring 11,400 images with disease segmentation masks and an additional 8,000 healthy plant images categorized by plant type. Extensive technical experiments validate the high quality of PlantSeg's annotations. This dataset not only allows researchers to evaluate their image classification methods but also provides a critical foundation for developing and benchmarking advanced plant disease segmentation algorithms.
This study focuses on identifying the most effective pre-trained model for land use classification in onboard satellite processing, emphasizing achieving high accuracy, computational efficiency, and robustness against noisy data conditions commonly encountered during satellite-based inference. Through extensive experimentation, we compare the performance of traditional CNN-based, ResNet-based, and various pre-trained vision Transformer models. Our findings demonstrate that pre-trained Vision Transformer (ViT) models, particularly MobileViTV2 and EfficientViT-M2, outperform models trained from scratch in terms of accuracy and efficiency. These models achieve high performance with reduced computational requirements and exhibit greater resilience during inference under noisy conditions. While MobileViTV2 has excelled on clean validation data, EfficientViT-M2 has proved more robust when handling noise, making it the most suitable model for onboard satellite EO tasks. Our experimental results demonstrate that EfficientViT-M2 is the optimal choice for reliable and efficient RS-IC in satellite operations, achieving 98.76 % of accuracy, precision, and recall. Precisely, EfficientViT-M2 delivers the highest performance across all metrics, excels in training efficiency (1,000s) and inference time (10s), and demonstrates greater robustness (overall robustness score of 0.79). Consequently, EfficientViT-M2 consumes 63.93 % less power than MobileViTV2 (79.23 W) and 73.26 % less power than SwinTransformer (108.90 W). This highlights its significant advantage in energy efficiency.
The aim of this paper is threefold. We inform the AI practitioner about the human visual system with an extensive literature review; we propose a novel biologically motivated neural network for image classification; and, finally, we present a new plug-and-play module to model context awareness. We focus on the effect of incorporating circuit motifs found in biological brains to address visual recognition. Our convolutional architecture is inspired by the connectivity of human cortical and subcortical streams, and we implement bottom-up and top-down modulations that mimic the extensive afferent and efferent connections between visual and cognitive areas. Our Contextual Attention Block is simple and effective and can be integrated with any feed-forward neural network. It infers weights that multiply the feature maps according to their causal influence on the scene, modeling the co-occurrence of different objects in the image. We place our module at different bottlenecks to infuse a hierarchical context awareness into the model. We validated our proposals through image classification experiments on benchmark data and found a consistent improvement in performance and the robustness of the produced explanations via class activation. Our code is available at https://github.com/gianlucarloni/CoCoReco.
Spiking Neural Networks (SNNs) have emerged as a promising substitute for Artificial Neural Networks (ANNs) due to their advantages of fast inference and low power consumption. However, the lack of efficient training algorithms has hindered their widespread adoption. Even efficient ANN-SNN conversion methods necessitate quantized training of ANNs to enhance the effectiveness of the conversion, incurring additional training costs. To address these challenges, we propose an efficient ANN-SNN conversion framework with only inference scale complexity. The conversion framework includes a local threshold balancing algorithm, which enables efficient calculation of the optimal thresholds and fine-grained adjustment of the threshold value by channel-wise scaling. We also introduce an effective delayed evaluation strategy to mitigate the influence of the spike propagation delays. We demonstrate the scalability of our framework in typical computer vision tasks: image classification, semantic segmentation, object detection, and video classification. Our algorithm outperforms existing methods, highlighting its practical applicability and efficiency. Moreover, we have evaluated the energy consumption of the converted SNNs, demonstrating their superior low-power advantage compared to conventional ANNs. This approach simplifies the deployment of SNNs by leveraging open-source pre-trained ANN models, enabling fast, low-power inference with negligible performance reduction. Code is available at https://github.com/putshua/Inference-scale-ANN-SNN.
Momentum based optimizers are central to a wide range of machine learning applications. These typically rely on an Exponential Moving Average (EMA) of gradients, which decays exponentially the present contribution of older gradients. This accounts for gradients being local linear approximations which lose their relevance as the iterate moves along the loss landscape. This work questions the use of a single EMA to accumulate past gradients and empirically demonstrates how this choice can be sub-optimal: a single EMA cannot simultaneously give a high weight to the immediate past, and a non-negligible weight to older gradients. Building on this observation, we propose AdEMAMix, a simple modification of the Adam optimizer with a mixture of two EMAs to better take advantage of past gradients. Our experiments on language modeling and image classification show -- quite surprisingly -- that gradients can stay relevant for tens of thousands of steps. They help to converge faster, and often to lower minima: e.g., a $1.3$B parameter AdEMAMix LLM trained on $101$B tokens performs comparably to an AdamW model trained on $197$B tokens ($+95\%$). Moreover, our method significantly slows-down model forgetting during training. Our work motivates further exploration of different types of functions to leverage past gradients, beyond EMAs.
Research in efficient vision backbones is evolving into models that are a mixture of convolutions and transformer blocks. A smart combination of both, architecture-wise and component-wise is mandatory to excel in the speedaccuracy trade-off. Most publications focus on maximizing accuracy and utilize MACs (multiply accumulate operations) as an efficiency metric. The latter however often do not measure accurately how fast a model actually is due to factors like memory access cost and degree of parallelism. We analyzed common modules and architectural design choices for backbones not in terms of MACs, but rather in actual throughput and latency, as the combination of the latter two is a better representation of the efficiency of models in real applications. We applied the conclusions taken from that analysis to create a recipe for increasing hardware-efficiency in macro design. Additionally we introduce a simple slimmed-down version of MultiHead Self-Attention, that aligns with our analysis. We combine both macro and micro design to create a new family of hardware-efficient backbone networks called LowFormer. LowFormer achieves a remarkable speedup in terms of throughput and latency, while achieving similar or better accuracy than current state-of-the-art efficient backbones. In order to prove the generalizability of our hardware-efficient design, we evaluate our method on GPU, mobile GPU and ARM CPU. We further show that the downstream tasks object detection and semantic segmentation profit from our hardware-efficient architecture. Code and models are available at https://github.com/ altair199797/LowFormer.
Convolutional Neural Networks (CNNs) have made remarkable strides; however, they remain susceptible to vulnerabilities, particularly in the face of minor image perturbations that humans can easily recognize. This weakness, often termed as 'attacks', underscores the limited robustness of CNNs and the need for research into fortifying their resistance against such manipulations. This study introduces a novel Non-Uniform Illumination (NUI) attack technique, where images are subtly altered using varying NUI masks. Extensive experiments are conducted on widely-accepted datasets including CIFAR10, TinyImageNet, and CalTech256, focusing on image classification with 12 different NUI attack models. The resilience of VGG, ResNet, MobilenetV3-small and InceptionV3 models against NUI attacks are evaluated. Our results show a substantial decline in the CNN models' classification accuracy when subjected to NUI attacks, indicating their vulnerability under non-uniform illumination. To mitigate this, a defense strategy is proposed, including NUI-attacked images, generated through the new NUI transformation, into the training set. The results demonstrate a significant enhancement in CNN model performance when confronted with perturbed images affected by NUI attacks. This strategy seeks to bolster CNN models' resilience against NUI attacks.
Large pre-trained vision-language models, such as CLIP, have demonstrated state-of-the-art performance across a wide range of image classification tasks, without requiring retraining. Few-shot CLIP is competitive with existing specialized architectures that were trained on the downstream tasks. Recent research demonstrates that the performance of CLIP can be further improved using lightweight adaptation approaches. However, previous methods adapt different modalities of the CLIP model individually, ignoring the interactions and relationships between visual and textual representations. In this work, we propose Multi-Modal Adapter, an approach for Multi-Modal adaptation of CLIP. Specifically, we add a trainable Multi-Head Attention layer that combines text and image features to produce an additive adaptation of both. Multi-Modal Adapter demonstrates improved generalizability, based on its performance on unseen classes compared to existing adaptation methods. We perform additional ablations and investigations to validate and interpret the proposed approach.
This paper addresses the limitations of conventional vector quantization algorithms, particularly K-Means and its variant K-Means++, and investigates the Stochastic Quantization (SQ) algorithm as a scalable alternative for high-dimensional unsupervised and semi-supervised learning tasks. Traditional clustering algorithms often suffer from inefficient memory utilization during computation, necessitating the loading of all data samples into memory, which becomes impractical for large-scale datasets. While variants such as Mini-Batch K-Means partially mitigate this issue by reducing memory usage, they lack robust theoretical convergence guarantees due to the non-convex nature of clustering problems. In contrast, the Stochastic Quantization algorithm provides strong theoretical convergence guarantees, making it a robust alternative for clustering tasks. We demonstrate the computational efficiency and rapid convergence of the algorithm on an image classification problem with partially labeled data, comparing model accuracy across various ratios of labeled to unlabeled data. To address the challenge of high dimensionality, we employ a Triplet Network to encode images into low-dimensional representations in a latent space, which serve as a basis for comparing the efficiency of both the Stochastic Quantization algorithm and traditional quantization algorithms. Furthermore, we enhance the algorithm's convergence speed by introducing modifications with an adaptive learning rate.
In this paper, we introduce FC-KAN, a Kolmogorov-Arnold Network (KAN) that leverages combinations of popular mathematical functions such as B-splines, wavelets, and radial basis functions on low-dimensional data through element-wise operations. We explore several methods for combining the outputs of these functions, including sum, element-wise product, the addition of sum and element-wise product, representations of quadratic and cubic functions, concatenation, linear transformation of the concatenated output, and others. In our experiments, we compare FC-KAN with a multi-layer perceptron network (MLP) and other existing KANs, such as BSRBF-KAN, EfficientKAN, FastKAN, and FasterKAN, on the MNIST and Fashion-MNIST datasets. Two variants of FC-KAN, which use a combination of outputs from B-splines and Difference of Gaussians (DoG) and from B-splines and linear transformations in the form of a quadratic function, outperformed overall other models on the average of 5 independent training runs. We expect that FC-KAN can leverage function combinations to design future KANs. Our repository is publicly available at: https://github.com/hoangthangta/FC_KAN.
Accurate classification of weather conditions in images is essential for enhancing the performance of object detection and classification models under varying weather conditions. This paper presents a comprehensive study on classifying weather conditions in images into four categories: rainy, low light, haze, and clear. The motivation for this work stems from the need to improve the reliability and efficiency of automated systems, such as autonomous vehicles and surveillance, which must operate under diverse weather conditions. Misclassification of weather conditions can lead to significant performance degradation in these systems, making robust weather classification crucial. Utilizing the Support Vector Machine (SVM) algorithm, our approach leverages a robust set of features, including brightness, saturation, noise level, blur metric, edge strength, motion blur, Local Binary Patterns (LBP) mean and variance for radii 1, 2, and 3, edges mean and variance, and color histogram mean and variance for blue, green, and red channels. Our SVM-based method achieved a notable accuracy of 92.8%, surpassing typical benchmarks in the literature, which range from 80% to 90% for classical machine learning methods. While deep learning methods can achieve up to 94% accuracy, our approach offers a competitive advantage in terms of computational efficiency and real-time classification capabilities. Detailed analysis of each feature's contribution highlights the effectiveness of texture, color, and edge-related features in capturing the unique characteristics of different weather conditions. This research advances the state-of-the-art in weather image classification and provides insights into the critical features necessary for accurate weather condition differentiation, underscoring the potential of SVMs in practical applications where accuracy is paramount.
Spiking Neural Networks (SNNs) have emerged as a compelling, energy-efficient alternative to traditional Artificial Neural Networks (ANNs) for static image tasks such as image classification and segmentation. However, in the more complex video classification domain, SNN-based methods fall considerably short of ANN-based benchmarks due to the challenges in processing dense frame sequences. To bridge this gap, we propose ReSpike, a hybrid framework that synergizes the strengths of ANNs and SNNs to tackle action recognition tasks with high accuracy and low energy cost. By decomposing film clips into spatial and temporal components, i.e., RGB image Key Frames and event-like Residual Frames, ReSpike leverages ANN for learning spatial information and SNN for learning temporal information. In addition, we propose a multi-scale cross-attention mechanism for effective feature fusion. Compared to state-of-the-art SNN baselines, our ReSpike hybrid architecture demonstrates significant performance improvements (e.g., >30% absolute accuracy improvement on HMDB-51, UCF-101, and Kinetics-400). Furthermore, ReSpike achieves comparable performance with prior ANN approaches while bringing better accuracy-energy tradeoff.