The risk-controlling prediction sets (RCPS) framework is a general tool for transforming the output of any machine learning model to design a predictive rule with rigorous error rate control. The key idea behind this framework is to use labeled hold-out calibration data to tune a hyper-parameter that affects the error rate of the resulting prediction rule. However, the limitation of such a calibration scheme is that with limited hold-out data, the tuned hyper-parameter becomes noisy and leads to a prediction rule with an error rate that is often unnecessarily conservative. To overcome this sample-size barrier, we introduce a semi-supervised calibration procedure that leverages unlabeled data to rigorously tune the hyper-parameter without compromising statistical validity. Our procedure builds upon the prediction-powered inference framework, carefully tailoring it to risk-controlling tasks. We demonstrate the benefits and validity of our proposal through two real-data experiments: few-shot image classification and early time series classification.
Whole-slide images (WSIs) are critical for cancer diagnosis due to their ultra-high resolution and rich semantic content. However, their massive size and the limited availability of fine-grained annotations pose substantial challenges for conventional supervised learning. We propose DSAGL (Dual-Stream Attention-Guided Learning), a novel weakly supervised classification framework that combines a teacher-student architecture with a dual-stream design. DSAGL explicitly addresses instance-level ambiguity and bag-level semantic consistency by generating multi-scale attention-based pseudo labels and guiding instance-level learning. A shared lightweight encoder (VSSMamba) enables efficient long-range dependency modeling, while a fusion-attentive module (FASA) enhances focus on sparse but diagnostically relevant regions. We further introduce a hybrid loss to enforce mutual consistency between the two streams. Experiments on CIFAR-10, NCT-CRC, and TCGA-Lung datasets demonstrate that DSAGL consistently outperforms state-of-the-art MIL baselines, achieving superior discriminative performance and robustness under weak supervision.
선천성 심장병(CHD)은 배아 발달 과정에서 심장과 대혈관의 비정상적 발달로 인해 발생하는 비정상적인 해부학적 구조를 말한다. 전통적인 진단 방법은 특히 심장 해부학의 복잡성을 고려할 때 높은 정확도와 효율성을 달성하지 못하는 경우가 많다. 본 연구는 심초음파 영상을 이용한 CHD 이진 분류를 위한 특화된 다중 뷰 심층 학습 프레임워크를 제안한다. 5개 뷰를 포함하는 대규모 CHD 데이터셋으로 모델을 학습하여 다각도 영상 데이터를 통합할 수 있도록 했다. 프레임워크는 고급 특징 추출과 어텐션 메커니즘을 활용하여 진단 정밀도와 신뢰성을 향상시킨다. 저품질 영상을 처리하기 위해 불확실성 기반 의사결정 구성 요소도 통합하여 진단 결과를 개선한다. 실험 결과, 본 방법은 우리 데이터셋에서 최상위 수준의 성능을 달성하고 조기 CHD 검출을 위한 견고한 도구를 제공하여 임상 적용 가능성을 보여준다. 데이터셋과 소스 코드는 논문 게재 승인 후 공개될 예정이다.
대조적 언어-이미지 학습(CLIP)은 원격 탐사 비전-언어 이해의 핵심 패러다임이 되었다. 그러나 기존 원격 탐사 대조 학습 방법은 대부분 RGB 중심 CLIP 아키텍처에 기반하여 SAR, 다중분광 영상(MSI), 초분광 영상(HSI)과 같은 이질적 센서를 활용하기 어렵다. 이 한계를 해결하기 위해, 우리는 원격 탐사 비전-언어 모델링에서 다중 소스 센서 입력을 지원하는 end-to-end 대조 학습 프레임워크 OmniRSCLIP을 제안한다. 핵심 아이디어는 사전학습된 시각 지식을 훼손하지 않으면서 CLIP을 고정 RGB 입력 인터페이스를 넘어 확장하는 것이다. 이를 위해 OmniRSCLIP은 Spectral-Spatial Basis Decomposition(SSBD)을 도입하며, 이는 임의 채널 적응을 기저 재구성 문제로 정식화한다. 사전학습된 CLIP 패치 임베딩은 전이 가능한 공간 기저를 제공하고, 파장 조건 계수는 제약된 시각적 사 prior 공간 내에서 센서별 임베딩 커널을 span한다. 이 설계는 이질적 센서를 고정 채널 입력 공간으로 강제하지 않으면서 통합된 이미지-텍스트 의미 공간에서 이들을 정렬한다. 또한 스펙트럼 맥락 인식 마스크 기반 대조 학습 방식을 도입하여 모달리티별 중복 특징을 억제하고 세밀한 이미지-텍스트 정렬을 강화한다. 마지막으로 다중 모달 학습을 지원하기 위해 RGB, SAR, MSI, HSI를 아우르는 최초의 대규모 원격 탐사 이미지-텍스트 코퍼스 OmniRS5M을 구축한다. 검색, 제로샷 분류, 의미적 localization 실험에서 OmniRSCLIP은 강한 RGB 도메인 성능을 유지하면서 CLIP을 이질적 원격 탐사 모달리티로 효과적으로 확장함을 보여준다.
Fingerprint-based recognition has been widely deployed in various applications. However, current recognition systems are vulnerable to spoofing attacks which make use of an artificial replica of a fingerprint to deceive the sensors. In such scenarios, fingerprint liveness detection ensures the actual presence of a real legitimate fingerprint in contrast to a fake self-manufactured synthetic sample. In this paper, we propose a static software-based approach using quality features to detect the liveness in a fingerprint. We have extracted features from a single fingerprint image to overcome the issues faced in dynamic software-based approaches which require longer computational time and user cooperation. The proposed system extracts 8 sensor independent quality features on a local level containing minute details of the ridge-valley structure of real and fake fingerprints. These local quality features constitutes a 13-dimensional feature vector. The system is tested on a publically available dataset of LivDet 2009 competition. The experimental results exhibit supremacy of the proposed method over current state-of-the-art approaches providing least average classification error of 5.3% for LivDet 2009. Additionally, effectiveness of the best performing features over LivDet 2009 is evaluated on the latest LivDet 2015 dataset which contain fingerprints fabricated using unknown spoof materials. An average classification error rate of 4.22% is achieved in comparison with 4.49% obtained by the LivDet 2015 winner. Further, the proposed system utilizes a single fingerprint image, which results in faster implications and makes it more user-friendly.
Training large-scale image recognition models is computationally expensive. This raises the question of whether there might be simple ways to improve the test performance of an already trained model without having to re-train or fine-tune it with new data. Here, we show that, surprisingly, this is indeed possible. The key observation we make is that the layers of a deep network close to the output layer contain independent, easily extractable class-relevant information that is not contained in the output layer itself. We propose to extract this extra class-relevant information using a simple key-value cache memory to improve the classification performance of the model at test time. Our cache memory is directly inspired by a similar cache model previously proposed for language modeling (Grave et al., 2017). This cache component does not require any training or fine-tuning; it can be applied to any pre-trained model and, by properly setting only two hyper-parameters, leads to significant improvements in its classification performance. Improvements are observed across several architectures and datasets. In the cache component, using features extracted from layers close to the output (but not from the output layer itself) as keys leads to the largest improvements. Concatenating features from multiple layers to form keys can further improve performance over using single-layer features as keys. The cache component also has a regularizing effect, a simple consequence of which is that it substantially increases the robustness of models against adversarial attacks.
Neural architecture search (NAS) has emerged as a promising avenue for automatically designing task-specific neural networks. Existing NAS approaches require one complete search for each deployment specification of hardware or objective. This is a computationally impractical endeavor given the potentially large number of application scenarios. In this paper, we propose Neural Architecture Transfer (NAT) to overcome this limitation. NAT is designed to efficiently generate task-specific custom models that are competitive under multiple conflicting objectives. To realize this goal we learn task-specific supernets from which specialized subnets can be sampled without any additional training. The key to our approach is an integrated online transfer learning and many-objective evolutionary search procedure. A pre-trained supernet is iteratively adapted while simultaneously searching for task-specific subnets. We demonstrate the efficacy of NAT on 11 benchmark image classification tasks ranging from large-scale multi-class to small-scale fine-grained datasets. In all cases, including ImageNet, NATNets improve upon the state-of-the-art under mobile settings ($\leq$ 600M Multiply-Adds). Surprisingly, small-scale fine-grained datasets benefit the most from NAT. At the same time, the architecture search and transfer is orders of magnitude more efficient than existing NAS methods. Overall, the experimental evaluation indicates that, across diverse image classification tasks and computational objectives, NAT is an appreciably more effective alternative to conventional transfer learning of fine-tuning weights of an existing network architecture learned on standard datasets. Code is available at https://github.com/human-analysis/neural-architecture-transfer
Recent self-supervised contrastive learning methods greatly benefit from the Siamese structure that aims to minimizing distances between positive pairs. These methods usually apply random data augmentation to input images, expecting the augmented views of the same images to be similar and positively paired. However, random augmentation may overlook image semantic information and degrade the quality of augmented views in contrastive learning. This issue becomes more challenging in medical images since the abnormalities related to diseases can be tiny, and are easy to be corrupted (e.g., being cropped out) in the current scheme of random augmentation. In this work, we first demonstrate that, for widely-used X-ray images, the conventional augmentation prevalent in contrastive pre-training can affect the performance of the downstream diagnosis or classification tasks. Then, we propose a novel augmentation method, i.e., FocusContrast, to learn from radiologists' gaze in diagnosis and generate contrastive views for medical images with guidance from radiologists' visual attention. Specifically, we track the gaze movement of radiologists and model their visual attention when reading to diagnose X-ray images. The learned model can predict visual attention of the radiologists given a new input image, and further guide the attention-aware augmentation that hardly neglects the disease-related abnormalities. As a plug-and-play and framework-agnostic module, FocusContrast consistently improves state-of-the-art contrastive learning methods of SimCLR, MoCo, and BYOL by 4.0~7.0% in classification accuracy on a knee X-ray dataset.
Salient Object Detection (SOD) has traditionally relied on feature refinement modules that utilize the features of an ImageNet pre-trained backbone. However, this approach limits the possibility of pre-training the entire network because of the distinct nature of SOD and image classification. Additionally, the architecture of these backbones originally built for Image classification is sub-optimal for a dense prediction task like SOD. To address these issues, we propose a novel encoder-decoder-style neural network called SODAWideNet++ that is designed explicitly for SOD. Inspired by the vision transformers ability to attain a global receptive field from the initial stages, we introduce the Attention Guided Long Range Feature Extraction (AGLRFE) module, which combines large dilated convolutions and self-attention. Specifically, we use attention features to guide long-range information extracted by multiple dilated convolutions, thus taking advantage of the inductive biases of a convolution operation and the input dependency brought by self-attention. In contrast to the current paradigm of ImageNet pre-training, we modify 118K annotated images from the COCO semantic segmentation dataset by binarizing the annotations to pre-train the proposed model end-to-end. Further, we supervise the background predictions along with the foreground to push our model to generate accurate saliency predictions. SODAWideNet++ performs competitively on five different datasets while only containing 35% of the trainable parameters compared to the state-of-the-art models. The code and pre-computed saliency maps are provided at https://github.com/VimsLab/SODAWideNetPlusPlus.
Automated diagnosis of eczema using images acquired from digital camera can enable individuals to self-monitor their recovery. The process entails first segmenting out the eczema region from the image and then measuring the severity of eczema in the segmented region. The state-of-the-art methods for automated eczema diagnosis rely on deep neural networks such as convolutional neural network (CNN) and have shown impressive performance in accurately measuring the severity of eczema. However, these methods require massive volume of annotated data to train which can be hard to obtain. In this paper, we propose a self-supervised learning framework for automated eczema diagnosis under limited training data regime. Our framework consists of two stages: i) Segmentation, where we use an in-context learning based algorithm called SegGPT for few-shot segmentation of eczema region from the image; ii) Feature extraction and classification, where we extract DINO features from the segmented regions and feed it to a multi-layered perceptron (MLP) for 4-class classification of eczema severity. When evaluated on a dataset of annotated "in-the-wild" eczema images, we show that our method outperforms (Weighted F1: 0.67 $\pm$ 0.01) the state-of-the-art deep learning methods such as finetuned Resnet-18 (Weighted F1: 0.44 $\pm$ 0.16) and Vision Transformer (Weighted F1: 0.40 $\pm$ 0.22). Our results show that self-supervised learning can be a viable solution for automated skin diagnosis where labeled data is scarce.
Data is commonly stored in tabular format. Several fields of research are prone to small imbalanced tabular data. Supervised Machine Learning on such data is often difficult due to class imbalance. Synthetic data generation, i.e., oversampling, is a common remedy used to improve classifier performance. State-of-the-art linear interpolation approaches, such as LoRAS and ProWRAS can be used to generate synthetic samples from the convex space of the minority class to improve classifier performance in such cases. Deep generative networks are common deep learning approaches for synthetic sample generation, widely used for synthetic image generation. However, their scope on synthetic tabular data generation in the context of imbalanced classification is not adequately explored. In this article, we show that existing deep generative models perform poorly compared to linear interpolation based approaches for imbalanced classification problems on smaller tabular datasets. To overcome this, we propose a deep generative model, ConvGeN that combines the idea of convex space learning with deep generative models. ConvGeN learns the coefficients for the convex combinations of the minority class samples, such that the synthetic data is distinct enough from the majority class. Our benchmarking experiments demonstrate that our proposed model ConvGeN improves imbalanced classification on such small datasets, as compared to existing deep generative models, while being at-par with the existing linear interpolation approaches. Moreover, we discuss how our model can be used for synthetic tabular data generation in general, even outside the scope of data imbalance and thus, improves the overall applicability of convex space learning.
The primary objective of domain adaptation methods is to transfer knowledge from a source domain to a target domain that has similar but different data distributions. Thus, in order to correctly classify the unlabeled target domain samples, the standard approach is to learn a common representation for both source and target domain, thereby indirectly addressing the problem of learning a classifier in the target domain. However, such an approach does not address the task of classification in the target domain directly. In contrast, we propose an approach that directly addresses the problem of learning a classifier in the unlabeled target domain. In particular, we train a classifier to correctly classify the training samples while simultaneously classifying the samples in the target domain in an unsupervised manner. The corresponding model is referred to as Discriminative Encoding for Domain Adaptation (DEDA). We show that this simple approach for performing unsupervised domain adaptation is indeed quite powerful. Our method achieves state of the art results in unsupervised adaptation tasks on various image classification benchmarks. We also obtained state of the art performance on domain adaptation in Amazon reviews sentiment classification dataset. We perform additional experiments when the source data has less labeled examples and also on zero-shot domain adaptation task where no target domain samples are used for training.
The field of deep learning is evolving in different directions, with still the need for more efficient training strategies. In this work, we present a novel and robust training scheme that integrates visual explanation techniques in the learning process. Unlike the attention mechanisms that focus on the relevant parts of images, we aim to improve the robustness of the model by making it pay attention to other regions as well. Broadly speaking, the idea is to distract the classifier in the learning process to force it to focus not only on relevant regions but also on those that, a priori, are not so informative for the discrimination of the class. We tested the proposed approach by embedding it into the learning process of a convolutional neural network for the analysis and classification of two well-known datasets, namely Stanford cars and FGVC-Aircraft. Furthermore, we evaluated our model on a real-case scenario for the classification of egocentric images, allowing us to obtain relevant information about peoples' lifestyles. In particular, we work on the challenging EgoFoodPlaces dataset, achieving state-of-the-art results with a lower level of complexity. The obtained results indicate the suitability of our proposed training scheme for image classification, improving the robustness of the final model.
Representation knowledge distillation aims at transferring rich information from one model to another. Common approaches for representation distillation mainly focus on the direct minimization of distance metrics between the models' embedding vectors. Such direct methods may be limited in transferring high-order dependencies embedded in the representation vectors, or in handling the capacity gap between the teacher and student models. Moreover, in standard knowledge distillation, the teacher is trained without awareness of the student's characteristics and capacity. In this paper, we explore two mechanisms for enhancing representation distillation using classifier sharing between the teacher and student. We first investigate a simple scheme where the teacher's classifier is connected to the student backbone, acting as an additional classification head. Then, we propose a student-aware mechanism that asks to tailor the teacher model to a student with limited capacity by training the teacher with a temporary student's head. We analyze and compare these two mechanisms and show their effectiveness on various datasets and tasks, including image classification, fine-grained classification, and face verification. In particular, we achieve state-of-the-art results for face verification on the IJB-C dataset for a MobileFaceNet model: TAR@(FAR=1e-5)=93.7\%. Code is available at https://github.com/Alibaba-MIIL/HeadSharingKD.
We devise deep nearest centroids (DNC), a conceptually elegant yet surprisingly effective network for large-scale visual recognition, by revisiting Nearest Centroids, one of the most classic and simple classifiers. Current deep models learn the classifier in a fully parametric manner, ignoring the latent data structure and lacking simplicity and explainability. DNC instead conducts nonparametric, case-based reasoning; it utilizes sub-centroids of training samples to describe class distributions and clearly explains the classification as the proximity of test data and the class sub-centroids in the feature space. Due to the distance-based nature, the network output dimensionality is flexible, and all the learnable parameters are only for data embedding. That means all the knowledge learnt for ImageNet classification can be completely transferred for pixel recognition learning, under the "pre-training and fine-tuning" paradigm. Apart from its nested simplicity and intuitive decision-making mechanism, DNC can even possess ad-hoc explainability when the sub-centroids are selected as actual training images that humans can view and inspect. Compared with parametric counterparts, DNC performs better on image classification (CIFAR-10, ImageNet) and greatly boots pixel recognition (ADE20K, Cityscapes), with improved transparency and fewer learnable parameters, using various network architectures (ResNet, Swin) and segmentation models (FCN, DeepLabV3, Swin). We feel this work brings fundamental insights into related fields.
Supervised classification methods have been widely utilized for the quality assurance of the advanced manufacturing process, such as additive manufacturing (AM) for anomaly (defects) detection. However, since abnormal states (with defects) occur much less frequently than normal ones (without defects) in a manufacturing process, the number of sensor data samples collected from a normal state is usually much more than that from an abnormal state. This issue causes imbalanced training data for classification analysis, thus deteriorating the performance of detecting abnormal states in the process. It is beneficial to generate effective artificial sample data for the abnormal states to make a more balanced training set. To achieve this goal, this paper proposes a novel data augmentation method based on a generative adversarial network (GAN) using additive manufacturing process image sensor data. The novelty of our approach is that a standard GAN and classifier are jointly optimized with techniques to stabilize the learning process of standard GAN. The diverse and high-quality generated samples provide balanced training data to the classifier. The iterative optimization between GAN and classifier provides the high-performance classifier. The effectiveness of the proposed method is validated by both open-source data and real-world case studies in polymer and metal AM processes.
Fine-grained visual classification (FGVC) is a challenging computer vision problem, where the task is to automatically recognise objects from subordinate categories. One of its main difficulties is capturing the most discriminative inter-class variances among visually similar classes. Recently, methods with Vision Transformer (ViT) have demonstrated noticeable achievements in FGVC, generally by employing the self-attention mechanism with additional resource-consuming techniques to distinguish potentially discriminative regions while disregarding the rest. However, such approaches may struggle to effectively focus on truly discriminative regions due to only relying on the inherent self-attention mechanism, resulting in the classification token likely aggregating global information from less-important background patches. Moreover, due to the immense lack of the datapoints, classifiers may fail to find the most helpful inter-class distinguishing features, since other unrelated but distinctive background regions may be falsely recognised as being valuable. To this end, we introduce a simple yet effective Salient Mask-Guided Vision Transformer (SM-ViT), where the discriminability of the standard ViT`s attention maps is boosted through salient masking of potentially discriminative foreground regions. Extensive experiments demonstrate that with the standard training procedure our SM-ViT achieves state-of-the-art performance on popular FGVC benchmarks among existing ViT-based approaches while requiring fewer resources and lower input image resolution.
Synthetic Aperture Radar (SAR) images are commonly utilized in military applications for automatic target recognition (ATR). Machine learning (ML) methods, such as Convolutional Neural Networks (CNN) and Graph Neural Networks (GNN), are frequently used to identify ground-based objects, including battle tanks, personnel carriers, and missile launchers. Determining the vehicle class, such as the BRDM2 tank, BMP2 tank, BTR60 tank, and BTR70 tank, is crucial, as it can help determine whether the target object is an ally or an enemy. While the ML algorithm provides feedback on the recognized target, the final decision is left to the commanding officers. Therefore, providing detailed information alongside the identified target can significantly impact their actions. This detailed information includes the SAR image features that contributed to the classification, the classification confidence, and the probability of the identified object being classified as a different object type or class. We propose a GNN-based ATR framework that provides the final classified class and outputs the detailed information mentioned above. This is the first study to provide a detailed analysis of the classification class, making final decisions more straightforward. Moreover, our GNN framework achieves an overall accuracy of 99.2\% when evaluated on the MSTAR dataset, improving over previous state-of-the-art GNN methods.
In recent years, deep neural networks (DNNs) have gained remarkable achievement in computer vision tasks, and the success of DNNs often depends greatly on the richness of data. However, the acquisition process of data and high-quality ground truth requires a lot of manpower and money. In the long, tedious process of data annotation, annotators are prone to make mistakes, resulting in incorrect labels of images, i.e., noisy labels. The emergence of noisy labels is inevitable. Moreover, since research shows that DNNs can easily fit noisy labels, the existence of noisy labels will cause significant damage to the model training process. Therefore, it is crucial to combat noisy labels for computer vision tasks, especially for classification tasks. In this survey, we first comprehensively review the evolution of different deep learning approaches for noisy label combating in the image classification task. In addition, we also review different noise patterns that have been proposed to design robust algorithms. Furthermore, we explore the inner pattern of real-world label noise and propose an algorithm to generate a synthetic label noise pattern guided by real-world data. We test the algorithm on the well-known real-world dataset CIFAR-10N to form a new real-world data-guided synthetic benchmark and evaluate some typical noise-robust methods on the benchmark.
In this contribution, we present a novel approach for segmenting laser radar (lidar) imagery into geometric time-height cloud locations with a fully convolutional network (FCN). We describe a semi-supervised learning method to train the FCN by: pre-training the classification layers of the FCN with image-level annotations, pre-training the entire FCN with the cloud locations of the MPLCMASK cloud mask algorithm, and fully supervised learning with hand-labeled cloud locations. We show the model achieves higher levels of cloud identification compared to the cloud mask algorithm implementation.