PyTorchKR Papers

“Image Classification”

총 18,385편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2109.07497공식 코드 · Papers with Code

Sign-MAML: Efficient Model-Agnostic Meta-Learning by SignSGD

We propose a new computationally-efficient first-order algorithm for Model-Agnostic Meta-Learning (MAML). The key enabling technique is to interpret MAML as a bilevel optimization (BLO) problem and leverage the sign-based SGD(signSGD) as a lower-level optimizer of BLO. We show that MAML, through the lens of signSGD-oriented BLO, naturally yields an alternating optimization scheme that just requires first-order gradients of a learned meta-model. We term the resulting MAML algorithm Sign-MAML. Compared to the conventional first-order MAML (FO-MAML) algorithm, Sign-MAML is theoretically-grounded as it does not impose any assumption on the absence of second-order derivatives during meta training. In practice, we show that Sign-MAML outperforms FO-MAML in various few-shot image classification tasks, and compared to MAML, it achieves a much more graceful tradeoff between classification accuracy and computation efficiency.

arXiv:2403.09318

A Hierarchical Fused Quantum Fuzzy Neural Network for Image Classification

Neural network is a powerful learning paradigm for data feature learning in the era of big data. However, most neural network models are deterministic models that ignore the uncertainty of data. Fuzzy neural networks are proposed to address this problem. FDNN is a hierarchical deep neural network that derives information from both fuzzy and neural representations, the representations are then fused to form representation to be classified. FDNN perform well on uncertain data classification tasks. In this paper, we proposed a novel hierarchical fused quantum fuzzy neural network (HQFNN). Different from classical FDNN, HQFNN uses quantum neural networks to learn fuzzy membership functions in fuzzy neural network. We conducted simulated experiment on two types of datasets (Dirty-MNIST and 15-Scene), the results show that the proposed model can outperform several existing methods. In addition, we demonstrate the robustness of the proposed quantum circuit.

arXiv:2403.19983

A multi-stage semi-supervised learning for ankle fracture classification on CT images

Because of the complicated mechanism of ankle injury, it is very difficult to diagnose ankle fracture in clinic. In order to simplify the process of fracture diagnosis, an automatic diagnosis model of ankle fracture was proposed. Firstly, a tibia-fibula segmentation network is proposed for the joint tibiofibular region of the ankle joint, and the corresponding segmentation dataset is established on the basis of fracture data. Secondly, the image registration method is used to register the bone segmentation mask with the normal bone mask. Finally, a semi-supervised classifier is constructed to make full use of a large number of unlabeled data to classify ankle fractures. Experiments show that the proposed method can segment fractures with fracture lines accurately and has better performance than the general method. At the same time, this method is superior to classification network in several indexes.

arXiv:2408.02676공식 코드 · Papers with Code

On Biases in a UK Biobank-based Retinal Image Classification Model

Recent work has uncovered alarming disparities in the performance of machine learning models in healthcare. In this study, we explore whether such disparities are present in the UK Biobank fundus retinal images by training and evaluating a disease classification model on these images. We assess possible disparities across various population groups and find substantial differences despite strong overall performance of the model. In particular, we discover unfair performance for certain assessment centres, which is surprising given the rigorous data standardisation protocol. We compare how these differences emerge and apply a range of existing bias mitigation methods to each one. A key insight is that each disparity has unique properties and responds differently to the mitigation methods. We also find that these methods are largely unable to enhance fairness, highlighting the need for better bias mitigation methods tailored to the specific type of bias.

arXiv:2409.08518공식 코드 · Papers with Code

Anytime Continual Learning for Open Vocabulary Classification

We propose an approach for anytime continual learning (AnytimeCL) for open vocabulary image classification. The AnytimeCL problem aims to break away from batch training and rigid models by requiring that a system can predict any set of labels at any time and efficiently update and improve when receiving one or more training samples at any time. Despite the challenging goal, we achieve substantial improvements over recent methods. We propose a dynamic weighting between predictions of a partially fine-tuned model and a fixed open vocabulary model that enables continual improvement when training samples are available for a subset of a task's labels. We also propose an attention-weighted PCA compression of training features that reduces storage and computation with little impact to model accuracy. Our methods are validated with experiments that test flexibility of learning and inference. Code is available at https://github.com/jessemelpolio/AnytimeCL.

arXiv:2506.20305

Learning Moderately Input-Sensitive Functions: A Case Study in QR Code Decoding

The hardness of learning a function that attains a target task relates to its input-sensitivity. For example, image classification tasks are input-insensitive as minor corruptions should not affect the classification results, whereas arithmetic and symbolic computation, which have been recently attracting interest, are highly input-sensitive as each input variable connects to the computation results. This study presents the first learning-based Quick Response (QR) code decoding and investigates learning functions of medium sensitivity. Our experiments reveal that Transformers can successfully decode QR codes, even beyond the theoretical error-correction limit, by learning the structure of embedded texts. They generalize from English-rich training data to other languages and even random strings. Moreover, we observe that the Transformer-based QR decoder focuses on data bits while ignoring error-correction bits, suggesting a decoding mechanism distinct from standard QR code readers.

arXiv:2607.13826v1cs.CV

심층 학습을 이용한 췌장암 절제 가능성의 멀티모달 평가

Multimodal Assessment of Pancreatic Cancer Resectability Using Deep Learning

췌관선암(pancreatic ductal adenocarcinoma, PDAC)의 절제 가능성을 정확히 판정하려면 CT 영상에서 종양이 주요 췌장 주변 혈관과 어떻게 상호작용하는지를 평가해야 하지만, 전문가 평가는 종종 상당한 변동성을 보인다. 본 연구에서는 3D 조영 CT와 구조화된 임상 정보를 공동으로 분석하여 환자를 National Comprehensive Cancer Network(NCCN)의 세 가지 절제 가능성 범주(즉시 절제 가능, 경계 절제 가능, 국소 진행)로 분류하는 완전 자동화 멀티모달 심층 학습 프레임워크를 제안한다. 이 접근법은 Swin-UNETR 백본을 사용하여 췌장, 종양, 혈관 구조의 보조 분할을 통해 해부학적으로 인식된 영상 표현을 얻는다. 이러한 특징은 일상적으로 수집되는 17개 변수에서 유도된 간결한 임상 임베딩과 융합되며, 경량 분류 헤드에 의해 처리된다. 모델 학습은 현재 종양 Dice 성능에 따라 분할과 분류 간 균형을 적응적으로 조절하는 동적 다중 과제 목적 함수에 의해 안내되며, 해부학적으로 정보를 담으면서도 판별력 있는 특징 표현을 유도한다.

arXiv:1703.09076

Active Convolution: Learning the Shape of Convolution for Image Classification

In recent years, deep learning has achieved great success in many computer vision applications. Convolutional neural networks (CNNs) have lately emerged as a major approach to image classification. Most research on CNNs thus far has focused on developing architectures such as the Inception and residual networks. The convolution layer is the core of the CNN, but few studies have addressed the convolution unit itself. In this paper, we introduce a convolution unit called the active convolution unit (ACU). A new convolution has no fixed shape, because of which we can define any form of convolution. Its shape can be learned through backpropagation during training. Our proposed unit has a few advantages. First, the ACU is a generalization of convolution; it can define not only all conventional convolutions, but also convolutions with fractional pixel coordinates. We can freely change the shape of the convolution, which provides greater freedom to form CNN structures. Second, the shape of the convolution is learned while training and there is no need to tune it by hand. Third, the ACU can learn better than a conventional unit, where we obtained the improvement simply by changing the conventional convolution to an ACU. We tested our proposed method on plain and residual networks, and the results showed significant improvement using our method on various datasets and architectures in comparison with the baseline.

arXiv:2307.03133공식 코드 · Papers with Code

Benchmarking Test-Time Adaptation against Distribution Shifts in Image Classification

Test-time adaptation (TTA) is a technique aimed at enhancing the generalization performance of models by leveraging unlabeled samples solely during prediction. Given the need for robustness in neural network systems when faced with distribution shifts, numerous TTA methods have recently been proposed. However, evaluating these methods is often done under different settings, such as varying distribution shifts, backbones, and designing scenarios, leading to a lack of consistent and fair benchmarks to validate their effectiveness. To address this issue, we present a benchmark that systematically evaluates 13 prominent TTA methods and their variants on five widely used image classification datasets: CIFAR-10-C, CIFAR-100-C, ImageNet-C, DomainNet, and Office-Home. These methods encompass a wide range of adaptation scenarios (e.g. online adaptation v.s. offline adaptation, instance adaptation v.s. batch adaptation v.s. domain adaptation). Furthermore, we explore the compatibility of different TTA methods with diverse network backbones. To implement this benchmark, we have developed a unified framework in PyTorch, which allows for consistent evaluation and comparison of the TTA methods across the different datasets and network architectures. By establishing this benchmark, we aim to provide researchers and practitioners with a reliable means of assessing and comparing the effectiveness of TTA methods in improving model robustness and generalization performance. Our code is available at https://github.com/yuyongcan/Benchmark-TTA.

arXiv:2404.08489

SpectralMamba: Efficient Mamba for Hyperspectral Image Classification

Recurrent neural networks and Transformers have recently dominated most applications in hyperspectral (HS) imaging, owing to their capability to capture long-range dependencies from spectrum sequences. However, despite the success of these sequential architectures, the non-ignorable inefficiency caused by either difficulty in parallelization or computationally prohibitive attention still hinders their practicality, especially for large-scale observation in remote sensing scenarios. To address this issue, we herein propose SpectralMamba -- a novel state space model incorporated efficient deep learning framework for HS image classification. SpectralMamba features the simplified but adequate modeling of HS data dynamics at two levels. First, in spatial-spectral space, a dynamical mask is learned by efficient convolutions to simultaneously encode spatial regularity and spectral peculiarity, thus attenuating the spectral variability and confusion in discriminative representation learning. Second, the merged spectrum can then be efficiently operated in the hidden state space with all parameters learned input-dependent, yielding selectively focused responses without reliance on redundant attention or imparallelizable recurrence. To explore the room for further computational downsizing, a piece-wise scanning mechanism is employed in-between, transferring approximately continuous spectrum into sequences with squeezed length while maintaining short- and long-term contextual profiles among hundreds of bands. Through extensive experiments on four benchmark HS datasets acquired by satellite-, aircraft-, and UAV-borne imagers, SpectralMamba surprisingly creates promising win-wins from both performance and efficiency perspectives.

arXiv:2411.00609

Tumor Location-weighted MRI-Report Contrastive Learning: A Framework for Improving the Explainability of Pediatric Brain Tumor Diagnosis

Despite the promising performance of convolutional neural networks (CNNs) in brain tumor diagnosis from magnetic resonance imaging (MRI), their integration into the clinical workflow has been limited. That is mainly due to the fact that the features contributing to a model's prediction are unclear to radiologists and hence, clinically irrelevant, i.e., lack of explainability. As the invaluable sources of radiologists' knowledge and expertise, radiology reports can be integrated with MRI in a contrastive learning (CL) framework, enabling learning from image-report associations, to improve CNN explainability. In this work, we train a multimodal CL architecture on 3D brain MRI scans and radiology reports to learn informative MRI representations. Furthermore, we integrate tumor location, salient to several brain tumor analysis tasks, into this framework to improve its generalizability. We then apply the learnt image representations to improve explainability and performance of genetic marker classification of pediatric Low-grade Glioma, the most prevalent brain tumor in children, as a downstream task. Our results indicate a Dice score of 31.1% between the model's attention maps and manual tumor segmentation (as an explainability measure) with test classification performance of 87.7%, significantly outperforming the baselines. These enhancements can build trust in our model among radiologists, facilitating its integration into clinical practices for more efficient tumor diagnosis.

arXiv:2607.10666v1cs.CV

소형 VLM을 이용한 퓨샷 산업 비전을 위한 답변 조건부 사고의 연쇄 지식 증류

Answer-Conditioned Chain-of-Thought Distillation for Few-Shot Industrial Vision with Small VLMs

제조업에서 AI 기반 시각 검사를 배포하는 것은 요구사항이 자주 변경되고, 새로운 결함 유형이 등장하며, 대규모 레이블 데이터셋을 구하기 어렵기 때문에 매우 어렵습니다. 우리는 최소한의 레이블 데이터만으로 소형 비전-언어 모델(VLM)을 새로운 산업 과제에 빠르게 적응시키기 위한 답변 조건부 사고의 연쇄(CoT) 지식 증류를 제안합니다. 최전선 VLM은 각 학습 이미지를 정답 레이블과 함께 입력받아 근거 있는 시각적 설명을 생성하며, 이후 30억 매개변수 모델이 LoRA를 통해 이러한 추론 보강 예제들로 미세조정됩니다. 정답에 조건화함으로써 모든 학습 추론이 올바른 결론을 향하도록 보장하는데, 이는 최전선 모델이 가장 어려운 과제에서 24.1%까지 낮은 점수를 기록하기 때문에 매우 중요합니다. 우리는 과제당 18개에서 30개의 레이블 이미지만 사용하여 세 가지 이미지 모달리티에 걸친 네 가지 산업 분류 과제에서 검증합니다. 과제당 4개의 시드(총 32회 학습 실행)에서, 우리의 방법은 16개의 모든 시드-과제 조합에서 직접 미세조정을 능가하며, 평균 +1.7에서 +4.4 퍼센트포인트 향상을 달성합니다. 통제된 동등 예산 실험을 통해 이 향상이 추가 학습 단계가 아닌 추론 품질에서 비롯됨을 확인합니다. 비조건부 기준선은 답변 조건화 없이는 잘못된 추론이 성능을 17.8 퍼센트포인트 저하시킴을 보여줍니다. 용접 방사선 사진 분류에서, 미세조정된 30억 매개변수 모델은 단 24개의 학습 이미지만으로 GPT-4.1을 10.0 퍼센트포인트 능가합니다.

arXiv:2610.00809v1cs.CV

너무 많은 토큰에 비용을 지불하고 있는가? 간단한 휴리스틱을 활용한 유효하고 비용 효율적인 멀티모달 LLM 주석

Paying for Too Many Tokens? Valid and Cost-Efficient Multimodal LLM Annotation with Simple Heuristics

비전-언어 모델(Vision-Language Models, VLMs)은 대규모로 비디오 주석 작업을 가능하게 하지만, 비용이 빠르게 누적됩니다: 초당 1프레임으로 일반적인 60초 분량의 숏폼 비디오를 처리하는 데에는 수백만 개의 토큰이 필요합니다. 비용을 줄이기 위해 연구자들은 프레임의 일부를 샘플링하거나, 비디오를 이미지 그리드로 압축하거나, 단일 모달리티만 사용하는 등의 휴리스틱에 의존합니다. 그러나 어떤 휴리스틱이 비용을 절감하는지, 그리고 그러한 휴리스틱이 해당 주석을 통해 도출되는 하위 결론(downstream conclusions)을 보존하는지는 여전히 명확하지 않습니다. 이러한 공백을 해소하기 위해, 우리는 숏폼 비디오를 사용하여 감성 분류와 주제 분류라는 두 가지 전산 사회과학(computational social science, CSS) 과제에 대해 이러한 휴리스틱들을 체계적으로 평가합니다. 우리는 선행 연구들이 일반적으로 별개로 다루어 온 세 가지 축, 즉 분류 정확도, 하위 추론(downstream inference)의 유효성, 비디오당 토큰 비용을 기준으로 각 구성(configuration)을 평가합니다. 첫째, 우리는 정확도와 유효성이 서로 다르게 나타난다는 것을 발견했습니다: 가장 높은 정확도를 보이는 구성이 잘못된 결론을 낳을 수 있습니다. 둘째, 모달리티의 가치는 보장되지 않습니다: 텍스트만으로도 강력한 성능을 낼 수 있으며, 이는 모달리티를 추가하는 것이 신호(signal)의 추가 없이 비용만 추가할 수 있음을 시사합니다. 마지막으로, 우리는 숏폼 비디오에 주석을 달 때 비용이 비디오 길이와 분리될 수 있음을 발견했습니다: 간단한 샷 전환 감지(shot-transition detection)를 통해 생성된 단일 $2\times8$ 이미지 그리드는 전체 비디오 이해 수준에 근접하며($κ$ 차이 .05 이내), 토큰 비용은 약 15%에 불과합니다. 이러한 발견을 바탕으로, 우리는 CSS 분야에서 비용을 고려한 VLM 주석 작업을 가능하게 하는 가이드라인을 도출합니다.

arXiv:2210.10654

POGD: Gradient Descent with New Stochastic Rules

There introduce Particle Optimized Gradient Descent (POGD), an algorithm based on the gradient descent but integrates the particle swarm optimization (PSO) principle to achieve the iteration. From the experiments, this algorithm has adaptive learning ability. The experiments in this paper mainly focus on the training speed to reach the target value and the ability to prevent the local minimum. The experiments in this paper are achieved by the convolutional neural network (CNN) image classification on the MNIST and cifar-10 datasets.

arXiv:2301.11015

Explore the Power of Dropout on Few-shot Learning

The generalization power of the pre-trained model is the key for few-shot deep learning. Dropout is a regularization technique used in traditional deep learning methods. In this paper, we explore the power of dropout on few-shot learning and provide some insights about how to use it. Extensive experiments on the few-shot object detection and few-shot image classification datasets, i.e., Pascal VOC, MS COCO, CUB, and mini-ImageNet, validate the effectiveness of our method.

arXiv:2309.06221

Use neural networks to recognize students' handwritten letters and incorrect symbols

Correcting students' multiple-choice answers is a repetitive and mechanical task that can be considered an image multi-classification task. Assuming possible options are 'abcd' and the correct option is one of the four, some students may write incorrect symbols or options that do not exist. In this paper, five classifications were set up - four for possible correct options and one for other incorrect writing. This approach takes into account the possibility of non-standard writing options.

arXiv:2504.02782공식 코드 · Papers with Code

GPT-ImgEval: A Comprehensive Benchmark for Diagnosing GPT4o in Image Generation

The recent breakthroughs in OpenAI's GPT4o model have demonstrated surprisingly good capabilities in image generation and editing, resulting in significant excitement in the community. This technical report presents the first-look evaluation benchmark (named GPT-ImgEval), quantitatively and qualitatively diagnosing GPT-4o's performance across three critical dimensions: (1) generation quality, (2) editing proficiency, and (3) world knowledge-informed semantic synthesis. Across all three tasks, GPT-4o demonstrates strong performance, significantly surpassing existing methods in both image generation control and output quality, while also showcasing exceptional knowledge reasoning capabilities. Furthermore, based on the GPT-4o's generated data, we propose a classification-model-based approach to investigate the underlying architecture of GPT-4o, where our empirical results suggest the model consists of an auto-regressive (AR) combined with a diffusion-based head for image decoding, rather than the VAR-like architectures. We also provide a complete speculation on GPT-4o's overall architecture. In addition, we conduct a series of analyses to identify and visualize GPT-4o's specific limitations and the synthetic artifacts commonly observed in its image generation. We also present a comparative study of multi-round image editing between GPT-4o and Gemini 2.0 Flash, and discuss the safety implications of GPT-4o's outputs, particularly their detectability by existing image forensic models. We hope that our work can offer valuable insight and provide a reliable benchmark to guide future research, foster reproducibility, and accelerate innovation in the field of image generation and beyond. The codes and datasets used for evaluating GPT-4o can be found at https://github.com/PicoTrex/GPT-ImgEval.

arXiv:1901.01677공식 코드 · Papers with Code

Image Super-Resolution as a Defense Against Adversarial Attacks

Convolutional Neural Networks have achieved significant success across multiple computer vision tasks. However, they are vulnerable to carefully crafted, human-imperceptible adversarial noise patterns which constrain their deployment in critical security-sensitive systems. This paper proposes a computationally efficient image enhancement approach that provides a strong defense mechanism to effectively mitigate the effect of such adversarial perturbations. We show that deep image restoration networks learn mapping functions that can bring off-the-manifold adversarial samples onto the natural image manifold, thus restoring classification towards correct classes. A distinguishing feature of our approach is that, in addition to providing robustness against attacks, it simultaneously enhances image quality and retains models performance on clean images. Furthermore, the proposed method does not modify the classifier or requires a separate mechanism to detect adversarial images. The effectiveness of the scheme has been demonstrated through extensive experiments, where it has proven a strong defense in gray-box settings. The proposed scheme is simple and has the following advantages: (1) it does not require any model training or parameter optimization, (2) it complements other existing defense mechanisms, (3) it is agnostic to the attacked model and attack type and (4) it provides superior performance across all popular attack algorithms. Our codes are publicly available at https://github.com/aamir-mustafa/super-resolution-adversarial-defense.

arXiv:2203.00047공식 코드 · Papers with Code

Local and Global GANs with Semantic-Aware Upsampling for Image Generation

In this paper, we address the task of semantic-guided image generation. One challenge common to most existing image-level generation methods is the difficulty in generating small objects and detailed local textures. To address this, in this work we consider generating images using local context. As such, we design a local class-specific generative network using semantic maps as guidance, which separately constructs and learns subgenerators for different classes, enabling it to capture finer details. To learn more discriminative class-specific feature representations for the local generation, we also propose a novel classification module. To combine the advantages of both global image-level and local class-specific generation, a joint generation network is designed with an attention fusion module and a dual-discriminator structure embedded. Lastly, we propose a novel semantic-aware upsampling method, which has a larger receptive field and can take far-away pixels that are semantically related for feature upsampling, enabling it to better preserve semantic consistency for instances with the same semantic labels. Extensive experiments on two image generation tasks show the superior performance of the proposed method. State-of-the-art results are established by large margins on both tasks and on nine challenging public benchmarks. The source code and trained models are available at https://github.com/Ha0Tang/LGGAN.

arXiv:1701.06190

A New Convolutional Network-in-Network Structure and Its Applications in Skin Detection, Semantic Segmentation, and Artifact Reduction

The inception network has been shown to provide good performance on image classification problems, but there are not much evidences that it is also effective for the image restoration or pixel-wise labeling problems. For image restoration problems, the pooling is generally not used because the decimated features are not helpful for the reconstruction of an image as the output. Moreover, most deep learning architectures for the restoration problems do not use dense prediction that need lots of training parameters. From these observations, for enjoying the performance of inception-like structure on the image based problems we propose a new convolutional network-in-network structure. The proposed network can be considered a modification of inception structure where pool projection and pooling layer are removed for maintaining the entire feature map size, and a larger kernel filter is added instead. Proposed network greatly reduces the number of parameters on account of removed dense prediction and pooling, which is an advantage, but may also reduce the receptive field in each layer. Hence, we add a larger kernel than the original inception structure for not increasing the depth of layers. The proposed structure is applied to typical image-to-image learning problems, i.e., the problems where the size of input and output are same such as skin detection, semantic segmentation, and compression artifacts reduction. Extensive experiments show that the proposed network brings comparable or better results than the state-of-the-art convolutional neural networks for these problems.