Recent advances in data augmentation enable one to translate images by learning the mapping between a source domain and a target domain. Existing methods tend to learn the distributions by training a model on a variety of datasets, with results evaluated largely in a subjective manner. Relatively few works in this area, however, study the potential use of image synthesis methods for recognition tasks. In this paper, we propose and explore the problem of image translation for data augmentation. We first propose a lightweight yet efficient model for translating texture to augment images based on a single input of source texture, allowing for fast training and testing, referred to as Single Image Texture Translation for data Augmentation (SITTA). Then we explore the use of augmented data in long-tailed and few-shot image classification tasks. We find the proposed augmentation method and workflow is capable of translating the texture of input data into a target domain, leading to consistently improved image recognition performance. Finally, we examine how SITTA and related image translation methods can provide a basis for a data-efficient, "augmentation engineering" approach to model training. Codes are available at https://github.com/Boyiliee/SITTA.
Recent powerful vision classifiers are biased towards textures, while shape information is overlooked by the models. A simple attempt by augmenting training images using the artistic style transfer method, called Stylized ImageNet, can reduce the texture bias. However, Stylized ImageNet approach has two drawbacks in fidelity and diversity. First, the generated images show low image quality due to the significant semantic gap betweeen natural images and artistic paintings. Also, Stylized ImageNet training samples are pre-computed before training, resulting in showing the lack of diversity for e…
Stony Brook University（石溪大学）；Memorial Sloan Kettering Cancer Center（纪念斯隆凯特琳癌症中心）
Automated analysis of optical colonoscopy (OC) video frames (to assist endoscopists during OC) is challenging due to variations in color, lighting, texture, and specular reflections. Previous methods either remove some of these variations via preprocessing (making pipelines cumbersome) or add diverse training data with annotations (but expensive and time-consuming). We present CLTS-GAN, a new deep learning model that gives fine control over color, lighting, texture, and specular reflection synthesis for OC video frames. We show that adding these colonoscopy-specific augmentations to the traini…
Shanghai Jiao Tong University（上海交通大学）；Shanghai United Imaging Intelligence Co., Ltd.（上海联影智能医疗科技有限公司）；ShanghaiTech University（上海科技大学）
Lung nodule detection in chest X-ray (CXR) images is common to early screening of lung cancers. Deep-learning-based Computer-Assisted Diagnosis (CAD) systems can support radiologists for nodule screening in CXR. However, it requires large-scale and diverse medical data with high-quality annotations to train such robust and accurate CADs. To alleviate the limited availability of such datasets, lung nodule synthesis methods are proposed for the sake of data augmentation. Nevertheless, previous methods lack the ability to generate nodules that are realistic with the size attribute desired by the …
We introduce style augmentation, a new form of data augmentation based on random style transfer, for improving the robustness of convolutional neural networks (CNN) over both classification and regression based tasks. During training, our style augmentation randomizes texture, contrast and color, while preserving shape and semantic content. This is accomplished by adapting an arbitrary style transfer network to perform style randomization, by sampling input style embeddings from a multivariate normal distribution instead of inferring them from a style image. In addition to standard classificat…
Evaluation of Randomization through Style Transfer for Enhanced Domain Generalization
German Cancer Research Center (DKFZ)（德国癌症研究中心（DKFZ））；German Cancer Consortium (DKTK), partner site Frankfurt/Mainz, a partnership between DKFZ and UCT Frankfurt-Marburg（德国癌症联盟（DKTK），法兰克福/美因茨合作站点，由DKFZ与UCT法兰克福-马尔堡合作）；Goethe University Frankfurt, Department of Computer Science（法兰克福大学计算机科学系）；The Hessian Center for Artificial Intelligence (hessian.AI)（黑森人工智能中心（hessian.AI））
Suboptimal generalization of machine learning models on unseen data is a key challenge which hampers the clinical applicability of such models to medical imaging. Although various methods such as domain adaptation and domain generalization have evolved to combat this challenge, learning robust and generalizable representations is core to medical image understanding, and continues to be a problem. Here, we propose STRAP (Style TRansfer Augmentation for histoPathology), a form of data augmentation based on random style transfer from non-medical style source such as artistic paintings, for learni…
Stony Brook University（石溪大学）；Rensselaer Polytechnic Institute（伦斯勒理工学院）；UT Southwestern Medical Center（德克萨斯大学西南医学中心）
Deep learning has shown great promise for CT image reconstruction, in particular to enable low dose imaging and integrated diagnostics. These merits, however, stand at great odds with the low availability of diverse image data which are needed to train these neural networks. We propose to overcome this bottleneck via a deep reinforcement learning (DRL) approach that is integrated with a style-transfer (ST) methodology, where the DRL generates the anatomical shapes and the ST synthesizes the texture detail. We show that our method bears high promise for generating novel and anatomically accurat…
Existing neural style transfer methods require reference style images to transfer texture information of style images to content images. However, in many practical situations, users may not have reference style images but still be interested in transferring styles by just imagining them. In order to deal with such applications, we propose a new framework that enables a style transfer `without' a style image, but only with a text description of the desired style. Using the pre-trained text-image embedding model of CLIP, we demonstrate the modulation of the style of content images only with a si…
Cornell University（康奈尔大学）；Delft University of Technology（代尔夫特理工大学）
A common strategy for improving model robustness is through data augmentations. Data augmentations encourage models to learn desired invariances, such as invariance to horizontal flipping or small changes in color. Recent work has shown that arbitrary style transfer can be used as a form of data augmentation to encourage invariance to textures by creating painting-like images from photographs. However, a stylized photograph is not quite the same as an artist-created painting. Artists depict perceptually meaningful cues in paintings so that humans can recognize salient components in scenes, an …
University of New South Wales（新南威尔士大学）；School of Minerals and Energy Resources Engineering（矿物与能源资源工程学院）
Digital Rock Imaging is constrained by detector hardware, and a trade-off between the image field of view (FOV) and the image resolution must be made. This can be compensated for with super resolution (SR) techniques that take a wide FOV, low resolution (LR) image, and super resolve a high resolution (HR), high FOV image. The Enhanced Deep Super Resolution Generative Adversarial Network (EDSRGAN) is trained on the Deep Learning Digital Rock Super Resolution Dataset, a diverse compilation 12000 of raw and processed uCT images. The network shows comparable performance of 50% to 70% reduction in …
Jadavpur University（贾达布尔大学）；Nanyang Technological University（南洋理工大学）；Cognizant Technology Solutions（高知特科技解决方案公司）；Indian Institute of Technology Roorkee（印度理工学院鲁尔基分校）
Binarization of degraded document images is an elementary step in most of the problems in document image analysis domain. The paper re-visits the binarization problem by introducing an adversarial learning approach. We construct a Texture Augmentation Network that transfers the texture element of a degraded reference document image to a clean binary image. In this way, the network creates multiple versions of the same textual content with various noisy textures, thus enlarging the available document binarization datasets. At last, the newly generated images are passed through a Binarization ne…
SkinGenBench: Generative Model and Preprocessing Effects for Synthetic Dermoscopic Augmentation in Melanoma Diagnosis
Alliance School of Advanced Computing, Alliance University, India（印度联盟大学高级计算学院）；Department of Computer Science and Communication, Østfold University College, 1783 Halden, Norway（挪威东福尔大学学院计算机科学与传播系）；Department of Dentistry and Oral Health, Aarhus University, 8000 Aarhus, Denmark（丹麦奥胡斯大学牙科与口腔健康系）
University of Granada（格拉纳达大学）；Technical University of Denmark (DTU)（丹麦技术大学）；Virginia Commonwealth University（弗吉尼亚联邦大学）
The recognition of coral species based on underwater texture images pose a significant difficulty for machine learning algorithms, due to the three following challenges embedded in the nature of this data: 1) datasets do not include information about the global structure of the coral; 2) several species of coral have very similar characteristics; and 3) defining the spatial borders between classes is difficult as many corals tend to appear together in groups. For this reason, the classification of coral species has always required an aid from a domain expert. The objective of this paper is to …
Data Augmentation for Surgical Scene Segmentation with Anatomy-Aware Diffusion Models
NCT/UCC Dresden（德累斯顿国家肿瘤疾病中心/德国癌症联盟德累斯顿分部）；DKFZ Heidelberg（德国癌症研究中心海德堡分部）；Faculty of Medicine & University Hospital Carl Gustav Carus TU Dresden（德累斯顿工业大学卡尔·古斯塔夫·卡鲁斯医学院及大学医院）；HZDR Dresden（德累斯顿亥姆霍兹中心）；The Centre for Tactile Internet with Human-in-the-Loop (CeTI)（以人为中心的触觉互联网中心）；Purdue University（普渡大学）；Indiana University（印第安纳大学）；University Hospital & Faculty of Medicine Carl Gustav Carus, TUD（德累斯顿工业大学卡尔·古斯塔夫·卡鲁斯医学院及大学医院）
Due to the limitation of available labeled data, medical image segmentation is a challenging task for deep learning. Traditional data augmentation techniques have been shown to improve segmentation network performances by optimizing the usage of few training examples. However, current augmentation approaches for segmentation do not tackle the strong texture bias of convolutional neural networks, observed in several studies. This work shows on the MoNuSeg dataset that style augmentation, which is already used in classification tasks, helps reducing texture over-fitting and improves segmentation…
University of Copenhagen（哥本哈根大学）；Gentofte Hospital（根措夫特医院）；ScreenPoint Medical
Purpose: Risk-stratified breast cancer screening might improve early detection and efficiency without comprising quality. However, modern mammography-based risk models do not ensure adaptation across vendor-domains and rely on cancer precursors, associated with short-term risk, which might limit long-term risk assessment. We report a cross-vendor mammographic texture model for long-term risk. Approach: The texture model was robustly trained using two systematically designed case-control datasets. Textural features, indicative of future breast cancer, were learned by excluding samples with diag…
The University of Tokyo（东京大学）；National Institute of Informatics（国立信息学研究所）；University of Cambridge（剑桥大学）；Cancer Research UK Cambridge Institute（英国癌症研究院剑桥研究所）；University of Milano-Bicocca（米兰比可卡大学）；Chubu University（中部大学）；Jikei University School of Medicine（东京慈惠会医科大学）；Kyushu University（九州大学）
Convolutional Neural Networks (CNNs) achieve excellent computer-assisted diagnosis with sufficient annotated training data. However, most medical imaging datasets are small and fragmented. In this context, Generative Adversarial Networks (GANs) can synthesize realistic/diverse additional training images to fill the data lack in the real image distribution; researchers have improved classification by augmenting data with noise-to-image (e.g., random noise samples to diverse pathological images) or image-to-image GANs (e.g., a benign image to a malignant one). Yet, no research has reported resul…
NAVER AI Lab（NAVER人工智能实验室）；Yonsei University（延世大学）
In generative adversarial networks, improving discriminators is one of the key components for generation performance. As image classifiers are biased toward texture and debiasing improves accuracy, we investigate 1) if the discriminators are biased, and 2) if debiasing the discriminators will improve generation performance. Indeed, we find empirical evidence that the discriminators are sensitive to the style (e.g., texture and color) of images. As a remedy, we propose feature statistics mixing regularization (FSMR) that encourages the discriminator's prediction to be invariant to the styles of…
Recent work has indicated that, unlike humans, ImageNet-trained CNNs tend to classify images by texture rather than by shape. How pervasive is this bias, and where does it come from? We find that, when trained on datasets of images with conflicting shape and texture, CNNs learn to classify by shape at least as easily as by texture. What factors, then, produce the texture bias in CNNs trained on ImageNet? Different unsupervised training objectives and different architectures have small but significant and largely independent effects on the level of texture bias. However, all objectives and arch…
SEDDI（SEDDI 公司）；Universidad Carlos III de Madrid（马德里卡洛斯三世大学）；Universidad Rey Juan Carlos（胡安·卡洛斯国王大学）
We present a deep learning-based method for propagating spatially-varying visual material attributes (e.g. texture maps or image stylizations) to larger samples of the same or similar materials. For training, we leverage images of the material taken under multiple illuminations and a dedicated data augmentation policy, making the transfer robust to novel illumination conditions and affine deformations. Our model relies on a supervised image-to-image translation framework and is agnostic to the transferred domain; we showcase a semantic segmentation, a normal map, and a stylization. Following a…
Recent studies have proven that DNNs, unlike human vision, tend to exploit texture information rather than shape. Such texture bias is one of the factors for the poor generalization performance of DNNs. We observe that the texture bias negatively affects not only in-domain generalization but also out-of-distribution generalization, i.e., Domain Generalization. Motivated by the observation, we propose a new framework to reduce the texture bias of a model by a novel optimization-based data augmentation, dubbed Stylized Dream. Our framework utilizes adaptive instance normalization (AdaIN) to augm…
Indian Institute of Science（印度科学学院）；Google Research（谷歌研究院）
Recent work has shown that deep vision models tend to be overly dependent on low-level or "texture" features, leading to poor generalization. Various data augmentation strategies have been proposed to overcome this so-called texture bias in DNNs. We propose a simple, lightweight adversarial augmentation technique that explicitly incentivizes the network to learn holistic shapes for accurate prediction in an object classification setting. Our augmentations superpose edgemaps from one image onto another image with shuffled patches, using a randomly determined mixing proportion, with the image la…
University of Oxford（牛津大学）；Facebook AI Research（Facebook人工智能研究院）
Deep networks for visual recognition are known to leverage "easy to recognise" portions of objects such as faces and distinctive texture patterns. The lack of a holistic understanding of objects may increase fragility and overfitting. In recent years, several papers have proposed to address this issue by means of occlusions as a form of data augmentation. However, successes have been limited to tasks such as weak localization and model interpretation, but no benefit was demonstrated on image classification on large-scale datasets. In this paper, we show that, by using a simple technique based …
University of Texas at Austin（德克萨斯大学奥斯汀分校）；UC Santa Barbara（加州大学圣塔芭芭拉分校）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）
The recent explosive interest on transformers has suggested their potential to become powerful "universal" models for computer vision tasks, such as classification, detection, and segmentation. While those attempts mainly study the discriminative models, we explore transformers on some more notoriously difficult vision tasks, e.g., generative adversarial networks (GANs). Our goal is to conduct the first pilot study in building a GAN completely free of convolutions, using only pure transformer-based architectures. Our vanilla GAN architecture, dubbed TransGAN, consists of a memory-friendly tran…
Medical Image Segmentation is a useful application for medical image analysis including detecting diseases and abnormalities in imaging modalities such as MRI, CT etc. Deep learning has proven to be promising for this task but usually has a low accuracy because of the lack of appropriate publicly available annotated or segmented medical datasets. In addition, the datasets that are available may have a different texture because of different dosage values or scanner properties than the images that need to be segmented. This paper presents a StyleGAN-driven approach for segmenting publicly availa…
MoireMix: A Formula-Based Data Augmentation for Improving Image Classification Robustness
National Institute of Advanced Industrial Science and Technology (AIST)（国立研究开发法人产业技术综合研究所）；Tokyo Denki University（东京电机大学）；Waseda University（早稻田大学）；Fundamental AI Lab, University of Technology Nuremberg（纽伦堡工业大学基础人工智能实验室）；Visual Geometry Group, University of Oxford（牛津大学视觉几何组）
We present a technique for zero-shot generation of a 3D model using only a target text prompt. Without any 3D supervision our method deforms the control shape of a limit subdivided surface along with its texture map and normal map to obtain a 3D asset that corresponds to the input text prompt and can be easily deployed into games or modeling applications. We rely only on a pre-trained CLIP model that compares the input text prompt with differentiably rendered images of our 3D model. While previous works have focused on stylization or required training of generative models we perform optimizati…
Data-Augmentation (DA) is known to improve performance across tasks and datasets. We propose a method to theoretically analyze the effect of DA and study questions such as: how many augmented samples are needed to correctly estimate the information encoded by that DA? How does the augmentation policy impact the final parameters of a model? We derive several quantities in close-form, such as the expectation and variance of an image, loss, and model's output under a given DA distribution. Those derivations open new avenues to quantify the benefits and limitations of DA. For example, we show that…
Mixup is a powerful data augmentation method that interpolates between two or more examples in the input or feature space and between the corresponding target labels. Many recent mixup methods focus on cutting and pasting two or more objects into one image, which is more about efficient processing than interpolation. However, how to best interpolate images is not well defined. In this sense, mixup has been connected to autoencoders, because often autoencoders "interpolate well", for instance generating an image that continuously deforms into another. In this work, we revisit mixup from the int…
Bosch Center for Artificial Intelligence（博世人工智能中心）；Max Planck Institute for Informatics（马克斯·普朗克信息学研究所）
Among the major remaining challenges for generative adversarial networks (GANs) is the capacity to synthesize globally and locally coherent images with object shapes and textures indistinguishable from real images. To target this issue we propose an alternative U-Net based discriminator architecture, borrowing the insights from the segmentation literature. The proposed U-Net based architecture allows to provide detailed per-pixel feedback to the generator while maintaining the global coherence of synthesized images, by providing the global image feedback as well. Empowered by the per-pixel res…
GigaMVS presents several challenges to existing Multi-View Stereo (MVS) algorithms for its large scale, complex occlusions, and gigapixel images. To address these problems, we first apply one of the state-of-the-art learning-based MVS methods, --MVSFormer, to overcome intractable scenarios such as textureless and reflections regions suffered by traditional PatchMatch methods, but it fails in a few large scenes' reconstructions. Moreover, traditional PatchMatch algorithms such as ACMMP, OpenMVS, and RealityCapture are leveraged to further improve the completeness in large scenes. Furthermore, t…
Domain generalization in semantic segmentation aims to alleviate the performance degradation on unseen domains through learning domain-invariant features. Existing methods diversify images in the source domain by adding complex or even abnormal textures to reduce the sensitivity to domain specific features. However, these approaches depend heavily on the richness of the texture bank, and training them can be time-consuming. In contrast to importing textures arbitrarily or augmenting styles randomly, we focus on the single source domain itself to achieve generalization. In this paper, we presen…
University of Central Florida（中佛罗里达大学）；Northwestern University（西北大学）；Mayo Clinic（梅奥诊所）；Sheikh Shakhbout Medical City（谢赫·沙赫布特医疗城）
Current data augmentation techniques and transformations are well suited for improving the size and quality of natural image datasets but are not yet optimized for medical imaging. We hypothesize that sub-optimal data augmentations can easily distort or occlude medical images, leading to false positives or negatives during patient diagnosis, prediction, or therapy/surgery evaluation. In our experimental results, we found that utilizing commonly used intensity-based data augmentation distorts the MRI scans and leads to texture information loss, thus negatively affecting the overall performance …
University of Leicester（莱斯特大学）；Ocean University of China（中国海洋大学）
Underwater object detection technique is of great significance for various applications in underwater the scenes. However, class imbalance issue is still an unsolved bottleneck for current underwater object detection algorithms. It leads to large precision discrepancies among different classes that the dominant classes with more training data achieve higher detection precisions while the minority classes with fewer training data achieves much lower detection precisions. In this paper, we propose a novel class-wise style augmentation (CWSA) algorithm to generate a class-balanced underwater data…
Graduate School of Information Sciences, Tohoku University（东北大学信息科学研究科）；Biometrics Research Laboratories, NEC Corporation（NEC公司生物识别研究所）
Although most fingerprint matching methods utilize minutia points and/or texture of fingerprint images as fingerprint features, the frequency spectrum is also a useful feature since a fingerprint is composed of ridge patterns with its inherent frequency band. We propose a novel CNN-based method for extracting fingerprint features from texture, minutiae, and frequency spectrum. In order to extract effective texture features from local regions around the minutiae, the minutia attention module is introduced to the proposed method. We also propose new data augmentation methods, which takes into ac…
Cartoonization is a task that renders natural photos into cartoon styles. Previous deep cartoonization methods only have focused on end-to-end translation, which may hinder editability. Instead, we propose a novel solution with editing features of texture and color based on the cartoon creation process. To do that, we design a model architecture to have separate decoders, texture and color, to decouple these attributes. In the texture decoder, we propose a texture controller, which enables a user to control stroke style and abstraction to generate diverse cartoon textures. We also introduce an…
Nanyang Technological University（南洋理工大学）；Institute of Engineering & Management（工程与管理学院）；Indian Institute of Technology Roorkee（印度理工学院鲁尔基分校）
With the large-scale explosion of images and videos over the internet, efficient hashing methods have been developed to facilitate memory and time efficient retrieval of similar images. However, none of the existing works uses hashing to address texture image retrieval mostly because of the lack of sufficiently large texture image databases. Our work addresses this problem by developing a novel deep learning architecture that generates binary hash codes for input texture images. For this, we first pre-train a Texture Synthesis Network (TSN) which takes a texture patch as input and outputs an e…
