PyTorchKR Papers

“Image Classification”

총 18,403편
태그·기간 등 상세 필터

카테고리 탐색

arXiv:2005.14302

Monocular Depth Estimators: Vulnerabilities and Attacks

Recent advancements of neural networks lead to reliable monocular depth estimation. Monocular depth estimated techniques have the upper hand over traditional depth estimation techniques as it only needs one image during inference. Depth estimation is one of the essential tasks in robotics, and monocular depth estimation has a wide variety of safety-critical applications like in self-driving cars and surgical devices. Thus, the robustness of such techniques is very crucial. It has been shown in recent works that these deep neural networks are highly vulnerable to adversarial samples for tasks like classification, detection and segmentation. These adversarial samples can completely ruin the output of the system, making their credibility in real-time deployment questionable. In this paper, we investigate the robustness of the most state-of-the-art monocular depth estimation networks against adversarial attacks. Our experiments show that tiny perturbations on an image that are invisible to the naked eye (perturbation attack) and corruption less than about 1% of an image (patch attack) can affect the depth estimation drastically. We introduce a novel deep feature annihilation loss that corrupts the hidden feature space representation forcing the decoder of the network to output poor depth maps. The white-box and black-box test compliments the effectiveness of the proposed attack. We also perform adversarial example transferability tests, mainly cross-data transferability.

arXiv:2006.01791공식 코드 · Papers with Code

SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularization

Advanced data augmentation strategies have widely been studied to improve the generalization ability of deep learning models. Regional dropout is one of the popular solutions that guides the model to focus on less discriminative parts by randomly removing image regions, resulting in improved regularization. However, such information removal is undesirable. On the other hand, recent strategies suggest to randomly cut and mix patches and their labels among training images, to enjoy the advantages of regional dropout without having any pointless pixel in the augmented images. We argue that such random selection strategies of the patches may not necessarily represent sufficient information about the corresponding object and thereby mixing the labels according to that uninformative patch enables the model to learn unexpected feature representation. Therefore, we propose SaliencyMix that carefully selects a representative image patch with the help of a saliency map and mixes this indicative patch with the target image, thus leading the model to learn more appropriate feature representation. SaliencyMix achieves the best known top-1 error of 21.26% and 20.09% for ResNet-50 and ResNet-101 architectures on ImageNet classification, respectively, and also improves the model robustness against adversarial perturbations. Furthermore, models that are trained with SaliencyMix help to improve the object detection performance. Source code is available at https://github.com/SaliencyMix/SaliencyMix.

arXiv:2312.13440공식 코드 · Papers with Code

MGAug: Multimodal Geometric Augmentation in Latent Spaces of Image Deformations

Geometric transformations have been widely used to augment the size of training images. Existing methods often assume a unimodal distribution of the underlying transformations between images, which limits their power when data with multimodal distributions occur. In this paper, we propose a novel model, Multimodal Geometric Augmentation (MGAug), that for the first time generates augmenting transformations in a multimodal latent space of geometric deformations. To achieve this, we first develop a deep network that embeds the learning of latent geometric spaces of diffeomorphic transformations (a.k.a. diffeomorphisms) in a variational autoencoder (VAE). A mixture of multivariate Gaussians is formulated in the tangent space of diffeomorphisms and serves as a prior to approximate the hidden distribution of image transformations. We then augment the original training dataset by deforming images using randomly sampled transformations from the learned multimodal latent space of VAE. To validate the efficiency of our model, we jointly learn the augmentation strategy with two distinct domain-specific tasks: multi-class classification on 2D synthetic datasets and segmentation on real 3D brain magnetic resonance images (MRIs). We also compare MGAug with state-of-the-art transformation-based image augmentation algorithms. Experimental results show that our proposed approach outperforms all baselines by significantly improved prediction accuracy. Our code is publicly available at https://github.com/tonmoy-hossain/MGAug.

arXiv:2411.17406공식 코드 · Papers with Code

CoA: Chain-of-Action for Generative Semantic Labels

Recent advances in vision-language models (VLM) have demonstrated remarkable capability in image classification. These VLMs leverage a predefined set of categories to construct text prompts for zero-shot reasoning. However, in more open-ended domains like autonomous driving, using a predefined set of labels becomes impractical, as the semantic label space is unknown and constantly evolving. Additionally, fixed embedding text prompts often tend to predict a single label (while in reality, multiple labels commonly exist per image). In this paper, we introduce CoA, an innovative Chain-of-Action (CoA) method that generates labels aligned with all contextually relevant features of an image. CoA is designed based on the observation that enriched and valuable contextual information improves generative performance during inference. Traditional vision-language models tend to output singular and redundant responses. Therefore, we employ a tailored CoA to alleviate this problem. We first break down the generative labeling task into detailed actions and construct an CoA leading to the final generative objective. Each action extracts and merges key information from the previous action and passes the enriched information as context to the next action, ultimately improving the VLM in generating comprehensive and accurate semantic labels. We assess the effectiveness of CoA through comprehensive evaluations on widely-used benchmark datasets and the results demonstrate significant improvements across key performance metrics.

arXiv:2611.17897v1cs.CV

See and Detect: 이미지 기반 오픈 어휘 다중 뷰 3D 객체 탐지

See and Detect: Image-Driven Open-Vocabulary Multi-View 3D Object Detection

최근의 오픈 어휘 다중 뷰 실내 3D 탐지 방법은 일반적으로 포인트 클라우드 기반 파이프라인을 따르며, 복원되거나 추정된 기하로부터 의사 3D 박스를 생성하고 투영된 영역으로부터 의미 레이블을 추론한다. 그러나 이러한 설계는 박스 생성을 포인트 클라우드의 완전성에 의존하게 하고 의미 학습을 국소적 외형에 한정하여, 3D 객체 위치 추정과 오픈 어휘 분류 성능을 떨어뜨린다. 본 논문에서는 다중 뷰 이미지를 3D 박스 및 의미 레이블 생성의 주된 근거로 사용하는 이미지 기반 프레임워크인 See and Detect를 제시한다. 박스 생성을 위해 먼저 개별 이미지에서 객체 후보를 발견하며, 이를 통해 포인트 클라우드에서 누락된 객체를 시각적 증거로부터 복구할 수 있다. 그러나 단일 뷰 예측은 잡음이 많고 중복되는 경우가 많다. 이를 해결하기 위해 뷰 간에 공간적으로 일관된 예측을 융합하는 다중 뷰 합의 박스 통합과, 객체를 부분적으로만 포함하는 박스의 영향을 줄이는 박스 품질 평가를 제안한다. 의미 레이블 생성을 위해서는 전체 이미지 맥락에서 의미 레이블을 질의하는 맥락 인식 의미 학습을 도입하여 더 정확한 의미 예측을 가능하게 한다. ScanNet과 ARKitScenes에서의 광범위한 실험은 제안 방법이 최고 수준의 오픈 어휘 다중 뷰 3D 탐지 성능을 달성하면서 기존 최고 성능 방법보다 500$\times$ 이상 빠르게 동작함을 보여 준다.

arXiv:2203.15536

BARC: Learning to Regress 3D Dog Shape from Images by Exploiting Breed Information

Our goal is to recover the 3D shape and pose of dogs from a single image. This is a challenging task because dogs exhibit a wide range of shapes and appearances, and are highly articulated. Recent work has proposed to directly regress the SMAL animal model, with additional limb scale parameters, from images. Our method, called BARC (Breed-Augmented Regression using Classification), goes beyond prior work in several important ways. First, we modify the SMAL shape space to be more appropriate for representing dog shape. But, even with a better shape model, the problem of regressing dog shape from an image is still challenging because we lack paired images with 3D ground truth. To compensate for the lack of paired data, we formulate novel losses that exploit information about dog breeds. In particular, we exploit the fact that dogs of the same breed have similar body shapes. We formulate a novel breed similarity loss consisting of two parts: One term encourages the shape of dogs from the same breed to be more similar than dogs of different breeds. The second one, a breed classification loss, helps to produce recognizable breed-specific shapes. Through ablation studies, we find that our breed losses significantly improve shape accuracy over a baseline without them. We also compare BARC qualitatively to WLDO with a perceptual study and find that our approach produces dogs that are significantly more realistic. This work shows that a-priori information about genetic similarity can help to compensate for the lack of 3D training data. This concept may be applicable to other animal species or groups of species. Our code is publicly available for research purposes at https://barc.is.tue.mpg.de/.

논문 ID: document/1014202

A Robust Feature Downsampling Module for Remote Sensing Visual Tasks

Remote-sensing (RS) images present unique challenges for computer vision (CV) due to lower resolution, smaller objects, and fewer features. Mainstream backbone networks show promising results for traditional visual tasks. However, they use convolution to reduce feature map dimensionality, which can result in information loss for small objects in RS images and decreased performance. To address this problem, we propose a new and universal downsampling module named robust feature downsampling (RFD). RFD fuses multiple feature maps extracted by different downsampling techniques, creating a more robust feature map with a complementary set of features. Leveraging this, we overcome the limitations of conventional convolutional downsampling, resulting in a more accurate and robust analysis of RS images. We develop two versions of the RFD module, shallow RFD (SRFD) and deep RFD (DRFD), tailored to adapt to different stages of feature capture and improve feature robustness. We replace the downsampling layers (DSL) of existing mainstream backbones with the RFD module and conduct comparative experiments on several public RS image datasets. The results show significant improvements compared to baseline approaches in RS image classification, object detection, and semantic segmentation. Specifically, our RFD module achieved an average performance gain of 1.5% on the NWPU-RESISC45 classification dataset without utilizing any additional pretraining data, resulting in state-of-the-art performance on this dataset. Moreover, in detection and segmentation tasks on dataset for object detection in aerial images (DOTA) and instance segmentation in aerial images dataset (iSAID), our RFD module outperforms the baseline approaches by 2%–7% when utilizing pretraining data from NWPU-RESISC45. These results highlight the value of the RFD module in enhancing the performance of RS visual tasks. The code is available at https://github.com/lwCVer/RFD.

arXiv:2506.00476

Towards Graph-Based Privacy-Preserving Federated Learning: ModelNet -- A ResNet-based Model Classification Dataset

Federated Learning (FL) has emerged as a powerful paradigm for training machine learning models across distributed data sources while preserving data locality. However, the privacy of local data is always a pivotal concern and has received a lot of attention in recent research on the FL regime. Moreover, the lack of domain heterogeneity and client-specific segregation in the benchmarks remains a critical bottleneck for rigorous evaluation. In this paper, we introduce ModelNet, a novel image classification dataset constructed from the embeddings extracted from a pre-trained ResNet50 model. First, we modify the CIFAR100 dataset into three client-specific variants, considering three domain heterogeneities (homogeneous, heterogeneous, and random). Subsequently, we train each client-specific subset of all three variants on the pre-trained ResNet50 model to save model parameters. In addition to multi-domain image data, we propose a new hypothesis to define the FL algorithm that can access the anonymized model parameters to preserve the local privacy in a more effective manner compared to existing ones. ModelNet is designed to simulate realistic FL settings by incorporating non-IID data distributions and client diversity design principles in the mainframe for both conventional and futuristic graph-driven FL algorithms. The three variants are ModelNet-S, ModelNet-D, and ModelNet-R, which are based on homogeneous, heterogeneous, and random data settings, respectively. To the best of our knowledge, we are the first to propose a cross-environment client-specific FL dataset along with the graph-based variant. Extensive experiments based on domain shifts and aggregation strategies show the effectiveness of the above variants, making it a practical benchmark for classical and graph-based FL research. The dataset and related code are available online.

arXiv:1607.05969

4D Cardiac Ultrasound Standard Plane Location by Spatial-Temporal Correlation

Echocardiography plays an important part in diagnostic aid in cardiac diseases. A critical step in echocardiography-aided diagnosis is to extract the standard planes since they tend to provide promising views to present different structures that are benefit to diagnosis. To this end, this paper proposes a spatial-temporal embedding framework to extract the standard view planes from 4D STIC (spatial-temporal image corre- lation) volumes. The proposed method is comprised of three stages, the frame smoothing, spatial-temporal embedding and final classification. In first stage, an L 0 smoothing filter is used to preprocess the frames that removes the noise and preserves the boundary. Then a compact repre- sentation is learned via embedding spatial and temporal features into a latent space in the supervised scheme considering both standard plane information and diagnosis result. In last stage, the learned features are fed into support vector machine to identify the standard plane. We eval- uate the proposed method on a 4D STIC volume dataset with 92 normal cases and 93 abnormal cases in three standard planes. It demonstrates that our method outperforms the baselines in both classification accuracy and computational efficiency.

arXiv:1502.03167

Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift

Training Deep Neural Networks is complicated by the fact that the distribution of each layer's inputs changes during training, as the parameters of the previous layers change. This slows down the training by requiring lower learning rates and careful parameter initialization, and makes it notoriously hard to train models with saturating nonlinearities. We refer to this phenomenon as internal covariate shift, and address the problem by normalizing layer inputs. Our method draws its strength from making normalization a part of the model architecture and performing the normalization for each training mini-batch. Batch Normalization allows us to use much higher learning rates and be less careful about initialization. It also acts as a regularizer, in some cases eliminating the need for Dropout. Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin. Using an ensemble of batch-normalized networks, we improve upon the best published result on ImageNet classification: reaching 4.9% top-5 validation error (and 4.8% test error), exceeding the accuracy of human raters.

arXiv:1811.10798공식 코드 · Papers with Code

Sequentially Aggregated Convolutional Networks

Modern deep networks generally implement a certain form of shortcut connections to alleviate optimization difficulties. However, we observe that such network topology alters the nature of deep networks. In many ways, these networks behave similarly to aggregated wide networks. We thus exploit the aggregation nature of shortcut connections at a finer architectural level and place them within wide convolutional layers. We end up with a sequentially aggregated convolutional (SeqConv) layer that combines the benefits of both wide and deep representations by aggregating features of various depths in sequence. The proposed SeqConv serves as a drop-in replacement of regular wide convolutional layers and thus could be handily integrated into any backbone network. We apply SeqConv to widely adopted backbones including ResNet and ResNeXt, and conduct experiments for image classification on public benchmark datasets. Our ResNet based network with a model size of ResNet-50 easily surpasses the performance of the 2.35$\times$ larger ResNet-152, while our ResNeXt based model sets a new state-of-the-art accuracy on ImageNet classification for networks with similar model complexity. The code and pre-trained models of our work are publicly available at https://github.com/GroupOfAlchemists/SeqConv.

arXiv:1902.00423

Do We Train on Test Data? Purging CIFAR of Near-Duplicates

The CIFAR-10 and CIFAR-100 datasets are two of the most heavily benchmarked datasets in computer vision and are often used to evaluate novel methods and model architectures in the field of deep learning. However, we find that 3.3% and 10% of the images from the test sets of these datasets have duplicates in the training set. These duplicates are easily recognizable by memorization and may, hence, bias the comparison of image recognition techniques regarding their generalization capability. To eliminate this bias, we provide the "fair CIFAR" (ciFAIR) dataset, where we replaced all duplicates in the test sets with new images sampled from the same domain. We then re-evaluate the classification performance of various popular state-of-the-art CNN architectures on these new test sets to investigate whether recent research has overfitted to memorizing data instead of learning abstract concepts. We find a significant drop in classification accuracy of between 9% and 14% relative to the original performance on the duplicate-free test set. The ciFAIR dataset and pre-trained models are available at https://cvjena.github.io/cifair/, where we also maintain a leaderboard.

arXiv:1902.06085

DC-AL GAN: Pseudoprogression and True Tumor Progression of Glioblastoma Multiform Image Classification Based on DCGAN and AlexNet

Pseudoprogression (PsP) occurs in 20-30% of patients with glioblastoma multiforme (GBM) after receiving the standard treatment. In the course of post-treatment magnetic resonance imaging (MRI), PsP exhibits similarities in shape and intensity to the true tumor progression (TTP) of GBM. So, these similarities pose challenges on the differentiation of these types of progression and hence the selection of the appropriate clinical treatment strategy. In this paper, we introduce DC-AL GAN, a novel feature learning method based on deep convolutional generative adversarial network (DCGAN) and AlexNet, to discriminate between PsP and TTP in MRI images. Due to the adversarial relationship between the generator and the discriminator of DCGAN, high-level discriminative features of PsP and TTP can be derived for the discriminator with AlexNet. Also, a feature fusion scheme is used to combine higher-layer features with lower-layer information, leading to more powerful features that are used for effectively discriminating between PsP and TTP. The experimental results show that DC-AL GAN achieves desirable PsP and TTP classification performance that is superior to other state-of-the-art methods.

arXiv:1907.12915공식 코드 · Papers with Code

Reg R-CNN: Lesion Detection and Grading under Noisy Labels

For the task of concurrently detecting and categorizing objects, the medical imaging community commonly adopts methods developed on natural images. Current state-of-the-art object detectors are comprised of two stages: the first stage generates region proposals, the second stage subsequently categorizes them. Unlike in natural images, however, for anatomical structures of interest such as tumors, the appearance in the image (e.g., scale or intensity) links to a malignancy grade that lies on a continuous ordinal scale. While classification models discard this ordinal relation between grades by discretizing the continuous scale to an unordered bag of categories, regression models are trained with distance metrics, which preserve the relation. This advantage becomes all the more important in the setting of label confusions on ambiguous data sets, which is the usual case with medical images. To this end, we propose Reg R-CNN, which replaces the second-stage classification model of a current object detector with a regression model. We show the superiority of our approach on a public data set with 1026 patients and a series of toy experiments. Code will be available at github.com/MIC-DKFZ/RegRCNN.

arXiv:1910.02624

Label-PEnet: Sequential Label Propagation and Enhancement Networks for Weakly Supervised Instance Segmentation

Weakly-supervised instance segmentation aims to detect and segment object instances precisely, given imagelevel labels only. Unlike previous methods which are composed of multiple offline stages, we propose Sequential Label Propagation and Enhancement Networks (referred as Label-PEnet) that progressively transform image-level labels to pixel-wise labels in a coarse-to-fine manner. We design four cascaded modules including multi-label classification, object detection, instance refinement and instance segmentation, which are implemented sequentially by sharing the same backbone. The cascaded pipeline is trained alternatively with a curriculum learning strategy that generalizes labels from high-level images to low-level pixels gradually with increasing accuracy. In addition, we design a proposal calibration module to explore the ability of classification networks to find key pixels that identify object parts, which serves as a post validation strategy running in the inverse order. We evaluate the efficiency of our Label-PEnet in mining instance masks on standard benchmarks: PASCAL VOC 2007 and 2012. Experimental results show that Label-PEnet outperforms the state-of-the-art algorithms by a clear margin, and obtains comparable performance even with the fully-supervised approaches.

arXiv:2108.00610공식 코드 · Papers with Code

Multiple Classifiers Based Maximum Classifier Discrepancy for Unsupervised Domain Adaptation

Adversarial training based on the maximum classifier discrepancy between two classifier structures has achieved great success in unsupervised domain adaptation tasks for image classification. The approach adopts the structure of two classifiers, though simple and intuitive, the learned classification boundary may not well represent the data property in the new domain. In this paper, we propose to extend the structure to multiple classifiers to further boost its performance. To this end, we develop a very straightforward approach to adding more classifiers. We employ the principle that the classifiers are different from each other to construct a discrepancy loss function for multiple classifiers. The proposed construction method of loss function makes it possible to add any number of classifiers to the original framework. The proposed approach is validated through extensive experimental evaluations. We demonstrate that, on average, adopting the structure of three classifiers normally yields the best performance as a trade-off between accuracy and efficiency. With minimum extra computational costs, the proposed approach can significantly improve the performance of the original algorithm. The source code of the proposed approach can be downloaded from \url{https://github.com/rucv/MMCD\_DA}.

arXiv:2211.11432공식 코드 · Papers with Code

MATE: Masked Autoencoders are Online 3D Test-Time Learners

Our MATE is the first Test-Time-Training (TTT) method designed for 3D data, which makes deep networks trained for point cloud classification robust to distribution shifts occurring in test data. Like existing TTT methods from the 2D image domain, MATE also leverages test data for adaptation. Its test-time objective is that of a Masked Autoencoder: a large portion of each test point cloud is removed before it is fed to the network, tasked with reconstructing the full point cloud. Once the network is updated, it is used to classify the point cloud. We test MATE on several 3D object classification datasets and show that it significantly improves robustness of deep networks to several types of corruptions commonly occurring in 3D point clouds. We show that MATE is very efficient in terms of the fraction of points it needs for the adaptation. It can effectively adapt given as few as 5% of tokens of each test sample, making it extremely lightweight. Our experiments show that MATE also achieves competitive performance by adapting sparsely on the test data, which further reduces its computational overhead, making it ideal for real-time applications.

arXiv:2307.05254공식 코드 · Papers with Code

OpenAL: An Efficient Deep Active Learning Framework for Open-Set Pathology Image Classification

Active learning (AL) is an effective approach to select the most informative samples to label so as to reduce the annotation cost. Existing AL methods typically work under the closed-set assumption, i.e., all classes existing in the unlabeled sample pool need to be classified by the target model. However, in some practical clinical tasks, the unlabeled pool may contain not only the target classes that need to be fine-grainedly classified, but also non-target classes that are irrelevant to the clinical tasks. Existing AL methods cannot work well in this scenario because they tend to select a large number of non-target samples. In this paper, we formulate this scenario as an open-set AL problem and propose an efficient framework, OpenAL, to address the challenge of querying samples from an unlabeled pool with both target class and non-target class samples. Experiments on fine-grained classification of pathology images show that OpenAL can significantly improve the query quality of target class samples and achieve higher performance than current state-of-the-art AL methods. Code is available at https://github.com/miccaiif/OpenAL.

arXiv:2407.10681공식 코드 · Papers with Code

GeoMix: Towards Geometry-Aware Data Augmentation

Mixup has shown considerable success in mitigating the challenges posed by limited labeled data in image classification. By synthesizing samples through the interpolation of features and labels, Mixup effectively addresses the issue of data scarcity. However, it has rarely been explored in graph learning tasks due to the irregularity and connectivity of graph data. Specifically, in node classification tasks, Mixup presents a challenge in creating connections for synthetic data. In this paper, we propose Geometric Mixup (GeoMix), a simple and interpretable Mixup approach leveraging in-place graph editing. It effectively utilizes geometry information to interpolate features and labels with those from the nearby neighborhood, generating synthetic nodes and establishing connections for them. We conduct theoretical analysis to elucidate the rationale behind employing geometry information for node Mixup, emphasizing the significance of locality enhancement-a critical aspect of our method's design. Extensive experiments demonstrate that our lightweight Geometric Mixup achieves state-of-the-art results on a wide variety of standard datasets with limited labeled data. Furthermore, it significantly improves the generalization capability of underlying GNNs across various challenging out-of-distribution generalization tasks. Our code is available at https://github.com/WtaoZhao/geomix.

arXiv:2607.18349v1cs.CV

픽셀에서 예후까지: 자동 4등급 백내장 중증도 분류를 위한 합성곱 신경망과 GLCM 특징 융합

From Pixel to Prognosis: Convolutional and GLCM Feature Fusion for Automated Four-Class Cataract Severity Classification

목적: 특수 안과 장비 없이 표준 소비자급 컬러 안구 사진으로 동작하는 저비용 자동 백내장 중증도 분류 시스템을 개발한다. 방법: 합성곱 신경망(CNN)의 심층 특징과 Hough 원 검출로 국소화한 동공 관심 영역(ROI)에서 추출한 다섯 가지 수작업 Gray-Level Co-occurrence Matrix(GLCM) 및 강도 기술자—평균 강도, 균일성, 표준편차, 대비, 에너지—를 융합하는 하이브리드 프레임워크를 설계했다. 방사형 기저 함수(RBF) 커널을 사용하는 다중 클래스 서포트 벡터 머신(SVM)이 각 영상을 네 가지 중증도 등급—정상, 미성숙, 성숙, 과성숙 백내장— 중 하나로 분류한다. 결과: 제안한 융합 시스템은 안과 클리닉에서 수집한 300장의 영상(클래스당 75장)으로 구성되고 안과 전문의가 라벨링한 테스트 세트에서 95.0% 정확도, 93.8% 민감도, 96.1% 특이도를 달성했으며, 텍스처만 사용한 방법(88.5%)과 CNN만 사용한 방법(91.3%) 기준선을 능가하고 최근 발표된 심층 학습 접근법을 넘어섰다. 결론: CNN-GLCM-SVM 융합 프레임워크는 GPU 가속이나 특수 카메라 없이도 경쟁력 있는 4등급 백내장 등급 분류를 제공하여, 자원이 제한된 환경에서 1차 의료 및 원격 의료 배포에 적합하다.