PyTorchKR Papers

“Image Classification”

총 18,385편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2408.07680공식 코드 · Papers with Code

A Spitting Image: Modular Superpixel Tokenization in Vision Transformers

Vision Transformer (ViT) architectures traditionally employ a grid-based approach to tokenization independent of the semantic content of an image. We propose a modular superpixel tokenization strategy which decouples tokenization and feature extraction; a shift from contemporary approaches where these are treated as an undifferentiated whole. Using on-line content-aware tokenization and scale- and shape-invariant positional embeddings, we perform experiments and ablations that contrast our approach with patch-based tokenization and randomized partitions as baselines. We show that our method significantly improves the faithfulness of attributions, gives pixel-level granularity on zero-shot unsupervised dense prediction tasks, while maintaining predictive performance in classification tasks. Our approach provides a modular tokenization framework commensurable with standard architectures, extending the space of ViTs to a larger class of semantically-rich models.

arXiv:2610.17412v1cs.CV

Bottlenecked semantic compression by representation reconstruction

Vision Transformers incur high computational costs when processing long visual token sequences. Existing reduction methods typically rank tokens independently and discard those deemed unimportant or redundant. We propose a reconstruction-guided framework for visual token condensation (ReConV), which summarizes the complete visual sequence with a compact Summary Bank (SMB) and retains tokens whose information is not adequately captured by the SMB, as indicated by their reconstruction residuals. At each compression layer, a lightweight position-conditioned decoder reconstructs the current tokens, leaving shared information compressed in the SMB while explicitly preserving poorly reconstructed details. Experiments on image classification and multimodal understanding demonstrate favorable performance-efficiency trade-offs: ReConV surpasses the uncompressed baseline with 48\% fewer GMACs on ImageNet and enables reusable image-level representations for efficient multi-turn inference. Code is available at TokenDrop.

arXiv:2011.11311

Uncovering the Bias in Facial Expressions

Over the past decades the machine and deep learning community has celebrated great achievements in challenging tasks such as image classification. The deep architecture of artificial neural networks together with the plenitude of available data makes it possible to describe highly complex relations. Yet, it is still impossible to fully capture what the deep learning model has learned and to verify that it operates fairly and without creating bias, especially in critical tasks, for instance those arising in the medical field. One example for such a task is the detection of distinct facial expressions, called Action Units, in facial images. Considering this specific task, our research aims to provide transparency regarding bias, specifically in relation to gender and skin color. We train a neural network for Action Unit classification and analyze its performance quantitatively based on its accuracy and qualitatively based on heatmaps. A structured review of our results indicates that we are able to detect bias. Even though we cannot conclude from our results that lower classification performance emerged solely from gender and skin color bias, these biases must be addressed, which is why we end by giving suggestions on how the detected bias can be avoided.

arXiv:2203.07815

Adversarial Counterfactual Augmentation: Application in Alzheimer's Disease Classification

Due to the limited availability of medical data, deep learning approaches for medical image analysis tend to generalise poorly to unseen data. Augmenting data during training with random transformations has been shown to help and became a ubiquitous technique for training neural networks. Here, we propose a novel adversarial counterfactual augmentation scheme that aims at finding the most \textit{effective} synthesised images to improve downstream tasks, given a pre-trained generative model. Specifically, we construct an adversarial game where we update the input \textit{conditional factor} of the generator and the downstream \textit{classifier} with gradient backpropagation alternatively and iteratively. This can be viewed as finding the `\textit{weakness}' of the classifier and purposely forcing it to \textit{overcome} its weakness via the generative model. To demonstrate the effectiveness of the proposed approach, we validate the method with the classification of Alzheimer's Disease (AD) as a downstream task. The pre-trained generative model synthesises brain images using age as conditional factor. Extensive experiments and ablation studies have been performed to show that the proposed approach improves classification performance and has potential to alleviate spurious correlations and catastrophic forgetting. Code will be released upon acceptance.

arXiv:2302.12317

Fact or Artifact? Revise Layer-wise Relevance Propagation on various ANN Architectures

Layer-wise relevance propagation (LRP) is a widely used and powerful technique to reveal insights into various artificial neural network (ANN) architectures. LRP is often used in the context of image classification. The aim is to understand, which parts of the input sample have highest relevance and hence most influence on the model prediction. Relevance can be traced back through the network to attribute a certain score to each input pixel. Relevance scores are then combined and displayed as heat maps and give humans an intuitive visual understanding of classification models. Opening the black box to understand the classification engine in great detail is essential for domain experts to gain trust in ANN models. However, there are pitfalls in terms of model-inherent artifacts included in the obtained relevance maps, that can easily be missed. But for a valid interpretation, these artifacts must not be ignored. Here, we apply and revise LRP on various ANN architectures trained as classifiers on geospatial and synthetic data. Depending on the network architecture, we show techniques to control model focus and give guidance to improve the quality of obtained relevance maps to separate facts from artifacts.

arXiv:2407.03009공식 코드 · Papers with Code

Model Guidance via Explanations Turns Image Classifiers into Segmentation Models

Heatmaps generated on inputs of image classification networks via explainable AI methods like Grad-CAM and LRP have been observed to resemble segmentations of input images in many cases. Consequently, heatmaps have also been leveraged for achieving weakly supervised segmentation with image-level supervision. On the other hand, losses can be imposed on differentiable heatmaps, which has been shown to serve for (1)~improving heatmaps to be more human-interpretable, (2)~regularization of networks towards better generalization, (3)~training diverse ensembles of networks, and (4)~for explicitly ignoring confounding input features. Due to the latter use case, the paradigm of imposing losses on heatmaps is often referred to as "Right for the right reasons". We unify these two lines of research by investigating semi-supervised segmentation as a novel use case for the Right for the Right Reasons paradigm. First, we show formal parallels between differentiable heatmap architectures and standard encoder-decoder architectures for image segmentation. Second, we show that such differentiable heatmap architectures yield competitive results when trained with standard segmentation losses. Third, we show that such architectures allow for training with weak supervision in the form of image-level labels and small numbers of pixel-level labels, outperforming comparable encoder-decoder models. Code is available at \url{https://github.com/Kainmueller-Lab/TW-autoencoder}.

arXiv:1904.00838

Learning More with Less: GAN-based Medical Image Augmentation

Convolutional Neural Network (CNN)-based accurate prediction typically requires large-scale annotated training data. In Medical Imaging, however, both obtaining medical data and annotating them by expert physicians are challenging; to overcome this lack of data, Data Augmentation (DA) using Generative Adversarial Networks (GANs) is essential, since they can synthesize additional annotated training data to handle small and fragmented medical images from various scanners--those generated images, realistic but completely novel, can further fill the real image distribution uncovered by the original dataset. As a tutorial, this paper introduces GAN-based Medical Image Augmentation, along with tricks to boost classification/object detection/segmentation performance using them, based on our experience and related work. Moreover, we show our first GAN-based DA work using automatic bounding box annotation, for robust CNN-based brain metastases detection on 256 x 256 MR images; GAN-based DA can boost 10% sensitivity in diagnosis with a clinically acceptable number of additional False Positives, even with highly-rough and inconsistent bounding boxes.

arXiv:1912.03406

Principal Component Properties of Adversarial Samples

Deep Neural Networks for image classification have been found to be vulnerable to adversarial samples, which consist of sub-perceptual noise added to a benign image that can easily fool trained neural networks, posing a significant risk to their commercial deployment. In this work, we analyze adversarial samples through the lens of their contributions to the principal components of each image, which is different than prior works in which authors performed PCA on the entire dataset. We investigate a number of state-of-the-art deep neural networks trained on ImageNet as well as several attacks for each of the networks. Our results demonstrate empirically that adversarial samples across several attacks have similar properties in their contributions to the principal components of neural network inputs. We propose a new metric for neural networks to measure their robustness to adversarial samples, termed the (k,p) point. We utilize this metric to achieve 93.36% accuracy in detecting adversarial samples independent of architecture and attack type for models trained on ImageNet.

arXiv:2101.07988

Semi-supervised Keypoint Localization

Knowledge about the locations of keypoints of an object in an image can assist in fine-grained classification and identification tasks, particularly for the case of objects that exhibit large variations in poses that greatly influence their visual appearance, such as wild animals. However, supervised training of a keypoint detection network requires annotating a large image dataset for each animal species, which is a labor-intensive task. To reduce the need for labeled data, we propose to learn simultaneously keypoint heatmaps and pose invariant keypoint representations in a semi-supervised manner using a small set of labeled images along with a larger set of unlabeled images. Keypoint representations are learnt with a semantic keypoint consistency constraint that forces the keypoint detection network to learn similar features for the same keypoint across the dataset. Pose invariance is achieved by making keypoint representations for the image and its augmented copies closer together in feature space. Our semi-supervised approach significantly outperforms previous methods on several benchmarks for human and animal body landmark localization.

arXiv:2203.00049

Towards Targeted Change Detection with Heterogeneous Remote Sensing Images for Forest Mortality Mapping

Several generic methods have recently been developed for change detection in heterogeneous remote sensing data, such as images from synthetic aperture radar (SAR) and multispectral radiometers. However, these are not well suited to detect weak signatures of certain disturbances of ecological systems. To resolve this problem we propose a new approach based on image-to-image translation and one-class classification (OCC). We aim to map forest mortality caused by an outbreak of geometrid moths in a sparsely forested forest-tundra ecotone using multisource satellite images. The images preceding and following the event are collected by Landsat-5 and RADARSAT-2, respectively. Using a recent deep learning method for change-aware image translation, we compute difference images in both satellites' respective domains. These differences are stacked with the original pre- and post-event images and passed to an OCC trained on a small sample from the targeted change class. The classifier produces a credible map of the complex pattern of forest mortality.

arXiv:2302.03668공식 코드 · Papers with Code

Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery

The strength of modern generative models lies in their ability to be controlled through text-based prompts. Typical "hard" prompts are made from interpretable words and tokens, and must be hand-crafted by humans. There are also "soft" prompts, which consist of continuous feature vectors. These can be discovered using powerful optimization methods, but they cannot be easily interpreted, re-used across models, or plugged into a text-based interface. We describe an approach to robustly optimize hard text prompts through efficient gradient-based optimization. Our approach automatically generates hard text-based prompts for both text-to-image and text-to-text applications. In the text-to-image setting, the method creates hard prompts for diffusion models, allowing API users to easily generate, discover, and mix and match image concepts without prior knowledge on how to prompt the model. In the text-to-text setting, we show that hard prompts can be automatically discovered that are effective in tuning LMs for classification.

arXiv:2607.02404v1cs.CV

Object-centric LeJEPA

LeJEPA를 객체 중심으로 확장한 Object-centric LeJEPA는 SAM이 제공하는 오프더셀프(off-the-shelf) 마스크를 학습 시 고정 입력으로 활용해 장면 분할과 표현 학습의 순환 의존성 문제를 우회한다. 의미론적 정렬 손실과 인스턴스 분리 손실을 결합한 결과, COCO의 10~100% 데이터 규모에서 이미지 수준 LeJEPA 대비 추적(DAVIS), 분류(ImageNet-1k), 분할(ADE20k), 재식별(NAVI) 등 다양한 다운스트림 태스크에서 우수한 성능을 달성했다.

arXiv:2607.27122v1cs.CV

소형 VLM의 다중 작업 학습을 통한 그라운딩 기반 위장관 내시경 VQA 연구

Towards Grounded GI Endoscopy VQA via Multi-Task Learning on Small VLMs

위장관(GI) 내시경 이미지 분석은 모델이 이미지에 대한 자유 형식의 임상적 질문에 답변해야 하는 시각적 질문 답변(VQA)으로 단일 레이블 분류에서 전환되고 있습니다. 최근의 비전-언어 모델(VLM)이 이 작업에서 유망한 답변 정확도를 달성하고 있지만, 임상적 도입을 위해서는 모델의 내부 표현이 답변의 근거가 되는 시각적 증거를 반영해야 합니다. 본 연구에서는 최소한의 추가 어노테이션으로 기존 VQA 데이터셋에서 보조 그라운딩(grounding) 및 설명 작업을 구성하는 간단한 다중 작업 미세조정 레시피를 제안합니다. 전문가가 주석을 추가한 용종 마스크를 직접 재사용하는 한편, Grad-CAM 위치 파악 기능이 있는 GI 도메인 사전학습 분류기를 통해 정답 마스크가 부족한 소견 범주에 대한 약한 지도(weak supervision)를 제공합니다. 세 가지 소형 VLM 백본을 Kvasir-VQA-x1에서 일치된 VQA 전용 및 다중 작업 레시피 하에 저순위 적응(LoRA)으로 미세조정하였으며, 분포 내(in-distribution) 및 분포 외(out-of-distribution) 데이터 모두에서 평가했을 때 답변 토큰과 관련 이미지 영역 간의 향상된 암시적 정렬과 함께 일관된 정확도 향상을 보임을 제시합니다.

arXiv:1901.08707

Surrogate Supervision for Medical Image Analysis: Effective Deep Learning From Limited Quantities of Labeled Data

We investigate the effectiveness of a simple solution to the common problem of deep learning in medical image analysis with limited quantities of labeled training data. The underlying idea is to assign artificial labels to abundantly available unlabeled medical images and, through a process known as surrogate supervision, pre-train a deep neural network model for the target medical image analysis task lacking sufficient labeled training data. In particular, we employ 3 surrogate supervision schemes, namely rotation, reconstruction, and colorization, in 4 different medical imaging applications representing classification and segmentation for both 2D and 3D medical images. 3 key findings emerge from our research: 1) pre-training with surrogate supervision is effective for small training sets; 2) deep models trained from initial weights pre-trained through surrogate supervision outperform the same models when trained from scratch, suggesting that pre-training with surrogate supervision should be considered prior to training any deep 3D models; 3) pre-training models in the medical domain with surrogate supervision is more effective than transfer learning from an unrelated domain (e.g., natural images), indicating the practical value of abundant unlabeled medical image data.

arXiv:2204.02404

Hospital-Agnostic Image Representation Learning in Digital Pathology

Whole Slide Images (WSIs) in digital pathology are used to diagnose cancer subtypes. The difference in procedures to acquire WSIs at various trial sites gives rise to variability in the histopathology images, thus making consistent diagnosis challenging. These differences may stem from variability in image acquisition through multi-vendor scanners, variable acquisition parameters, and differences in staining procedure; as well, patient demographics may bias the glass slide batches before image acquisition. These variabilities are assumed to cause a domain shift in the images of different hospitals. It is crucial to overcome this domain shift because an ideal machine-learning model must be able to work on the diverse sources of images, independent of the acquisition center. A domain generalization technique is leveraged in this study to improve the generalization capability of a Deep Neural Network (DNN), to an unseen histopathology image set (i.e., from an unseen hospital/trial site) in the presence of domain shift. According to experimental results, the conventional supervised-learning regime generalizes poorly to data collected from different hospitals. However, the proposed hospital-agnostic learning can improve the generalization considering the low-dimensional latent space representation visualization, and classification accuracy results.

arXiv:2209.06726

Efficient Unsupervised Learning for Plankton Images

Monitoring plankton populations in situ is fundamental to preserve the aquatic ecosystem. Plankton microorganisms are in fact susceptible of minor environmental perturbations, that can reflect into consequent morphological and dynamical modifications. Nowadays, the availability of advanced automatic or semi-automatic acquisition systems has been allowing the production of an increasingly large amount of plankton image data. The adoption of machine learning algorithms to classify such data may be affected by the significant cost of manual annotation, due to both the huge quantity of acquired data and the numerosity of plankton species. To address these challenges, we propose an efficient unsupervised learning pipeline to provide accurate classification of plankton microorganisms. We build a set of image descriptors exploiting a two-step procedure. First, a Variational Autoencoder (VAE) is trained on features extracted by a pre-trained neural network. We then use the learnt latent space as image descriptor for clustering. We compare our method with state-of-the-art unsupervised approaches, where a set of pre-defined hand-crafted features is used for clustering of plankton images. The proposed pipeline outperforms the benchmark algorithms for all the plankton datasets included in our analysis, providing better image embedding properties.

arXiv:2205.13092공식 코드 · Papers with Code

Dual-Perspective Semantic-Aware Representation Blending for Multi-Label Image Recognition with Partial Labels

Despite achieving impressive progress, current multi-label image recognition (MLR) algorithms heavily depend on large-scale datasets with complete labels, making collecting large-scale datasets extremely time-consuming and labor-intensive. Training the multi-label image recognition models with partial labels (MLR-PL) is an alternative way, in which merely some labels are known while others are unknown for each image. However, current MLP-PL algorithms rely on pre-trained image similarity models or iteratively updating the image classification models to generate pseudo labels for the unknown labels. Thus, they depend on a certain amount of annotations and inevitably suffer from obvious performance drops, especially when the known label proportion is low. To address this dilemma, we propose a dual-perspective semantic-aware representation blending (DSRB) that blends multi-granularity category-specific semantic representation across different images, from instance and prototype perspective respectively, to transfer information of known labels to complement unknown labels. Specifically, an instance-perspective representation blending (IPRB) module is designed to blend the representations of the known labels in an image with the representations of the corresponding unknown labels in another image to complement these unknown labels. Meanwhile, a prototype-perspective representation blending (PPRB) module is introduced to learn more stable representation prototypes for each category and blends the representation of unknown labels with the prototypes of corresponding labels, in a location-sensitive manner, to complement these unknown labels. Extensive experiments on the MS-COCO, Visual Genome, and Pascal VOC 2007 datasets show that the proposed DSRB consistently outperforms current state-of-the-art algorithms on all known label proportion settings.

arXiv:1811.07753

Contextual Face Recognition with a Nested-Hierarchical Nonparametric Identity Model

Current face recognition systems typically operate via classification into known identities obtained from supervised identity annotations. There are two problems with this paradigm: (1) current systems are unable to benefit from often abundant unlabelled data; and (2) they equate successful recognition with labelling a given input image. Humans, on the other hand, regularly perform identification of individuals completely unsupervised, recognising the identity of someone they have seen before even without being able to name that individual. How can we go beyond the current classification paradigm towards a more human understanding of identities? In previous work, we proposed an integrated Bayesian model that coherently reasons about the observed images, identities, partial knowledge about names, and the situational context of each observation. Here, we propose extensions of the contextual component of this model, enabling unsupervised discovery of an unbounded number of contexts for improved face recognition.

arXiv:1901.02413

Interpretable CNNs for Object Classification

This paper proposes a generic method to learn interpretable convolutional filters in a deep convolutional neural network (CNN) for object classification, where each interpretable filter encodes features of a specific object part. Our method does not require additional annotations of object parts or textures for supervision. Instead, we use the same training data as traditional CNNs. Our method automatically assigns each interpretable filter in a high conv-layer with an object part of a certain category during the learning process. Such explicit knowledge representations in conv-layers of CNN help people clarify the logic encoded in the CNN, i.e., answering what patterns the CNN extracts from an input image and uses for prediction. We have tested our method using different benchmark CNNs with various structures to demonstrate the broad applicability of our method. Experiments have shown that our interpretable filters are much more semantically meaningful than traditional filters.

arXiv:2003.06581

Semi-supervised Disentanglement with Independent Vector Variational Autoencoders

We aim to separate the generative factors of data into two latent vectors in a variational autoencoder. One vector captures class factors relevant to target classification tasks, while the other vector captures style factors relevant to the remaining information. To learn the discrete class features, we introduce supervision using a small amount of labeled data, which can simply yet effectively reduce the effort required for hyperparameter tuning performed in existing unsupervised methods. Furthermore, we introduce a learning objective to encourage statistical independence between the vectors. We show that (i) this vector independence term exists within the result obtained on decomposing the evidence lower bound with multiple latent vectors, and (ii) encouraging such independence along with reducing the total correlation within the vectors enhances disentanglement performance. Experiments conducted on several image datasets demonstrate that the disentanglement achieved via our method can improve classification performance and generation controllability.