“Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告

The purpose of this review is to discuss the development and the state of the art in dynamic testing techniques and dynamic mechanical behaviour of rock materials. The review begins by briefly introducing the history of rock dynamics and explaining the significance of studying these issues. Loading techniques commonly used for both intermediate and high strain rate tests and measurement techniques for dynamic stress and deformation are critically assessed in Sects. 2 and 3. In Sect. 4, methods of dynamic testing and estimation to obtain stress–strain curves at high strain rate are summarized, followed by an in-depth description of various dynamic mechanical properties (e.g. uniaxial and triaxial compressive strength, tensile strength, shear strength and fracture toughness) and corresponding fracture behaviour. Some influencing rock structural features (i.e. microstructure, size and shape) and testing conditions (i.e. confining pressure, temperature and water saturation) are considered, ending with some popular semi-empirical rate-dependent equations for the enhancement of dynamic mechanical properties. Section 5 discusses physical mechanisms of strain rate effects. Section 6 describes phenomenological and mechanically based rate-dependent constitutive models established from the knowledge of the stress–strain behaviour and physical mechanisms. Section 7 presents dynamic fracture criteria for quasi-brittle materials. Finally, a brief summary and some aspects of prospective research are presented.

/pdf/a-review-of-dynamic-experimental-techniques-and-mechanical-2kstp2wfg4.pdf

A Review of Dynamic Experimental Techniques and Mechanical Behaviour of Rock Materials

Human adaptability relies crucially on the ability to learn and merge knowledge both from supervised and unsupervised learning: the parents point out few important concepts, but then the children fill in the gaps on their own. This is particularly effective, because supervised learning can never be exhaustive and thus learning autonomously allows to discover invariances and regularities that help to generalize. In this paper we propose to apply a similar approach to the task of object recognition across domains: our model learns the semantic labels in a supervised fashion, and broadens its understanding of the data by learning from self-supervised signals how to solve a jigsaw puzzle on the same images. This secondary task helps the network to learn the concepts of spatial correlation while acting as a regularizer for the classification task. Multiple experiments on the PACS, VLCS, Office-Home and digits datasets confirm our intuition and show that this simple method outperforms previous domain generalization and adaptation solutions. An ablation study further illustrates the inner workings of our approach.

/pdf/domain-generalization-by-solving-jigsaw-puzzles-1mgngt3e9w.pdf

Domain Generalization by Solving Jigsaw Puzzles

Distribution shifts -- where the training distribution differs from the test distribution -- can substantially degrade the accuracy of machine learning (ML) systems deployed in the wild. Despite their ubiquity, these real-world distribution shifts are under-represented in the datasets widely used in the ML community today. To address this gap, we present WILDS, a curated collection of 8 benchmark datasets that reflect a diverse range of distribution shifts which naturally arise in real-world applications, such as shifts across hospitals for tumor identification; across camera traps for wildlife monitoring; and across time and location in satellite imaging and poverty mapping. On each dataset, we show that standard training results in substantially lower out-of-distribution than in-distribution performance, and that this gap remains even with models trained by existing methods for handling distribution shifts. This underscores the need for new training methods that produce models which are more robust to the types of distribution shifts that arise in practice. To facilitate method development, we provide an open-source package that automates dataset loading, contains default model architectures and hyperparameters, and standardizes evaluations. Code and leaderboards are available at this https URL.

WILDS: A Benchmark of in-the-Wild Distribution Shifts

The new developments in deep generative networks have significantly improve the quality and efficiency in generating realistically-looking fake face videos. In this work, we describe a new method to expose fake face videos generated with deep neural network models. Our method is based on detection of eye blinking in the videos, which is a physiological signal that is not well presented in the synthesized fake videos. Our method is evaluated over benchmarks of eye-blinking detection datasets and shows promising performance on detecting videos generated with DNN based software DeepFake.

In Ictu Oculi: Exposing AI Created Fake Videos by Detecting Eye Blinking

In this paper, we tackle the problem of domain generalization: how to learn a generalized feature representation for an "unseen" target domain by taking the advantage of multiple seen source-domain data. We present a novel framework based on adversarial autoencoders to learn a generalized latent feature representation across domains for domain generalization. To be specific, we extend adversarial autoencoders by imposing the Maximum Mean Discrepancy (MMD) measure to align the distributions among different domains, and matching the aligned distribution to an arbitrary prior distribution via adversarial feature learning. In this way, the learned feature representation is supposed to be universal to the seen source domains because of the MMD regularization, and is expected to generalize well on the target domain because of the introduction of the prior distribution. We proposed an algorithm to jointly train different components of our proposed framework. Extensive experiments on various vision tasks demonstrate that our proposed framework can learn better generalized features for the unseen target domain compared with state-of-the-art domain generalization methods.

/pdf/domain-generalization-with-adversarial-feature-learning-5e5o74l6ns.pdf

Domain Generalization with Adversarial Feature Learning

Face anti-spoofing (a.k.a. presentation attack detection) has recently emerged as an active topic with great significance for both academia and industry due to the rapidly increasing demand in user authentication on mobile phones, PCs, tablets, and so on. Recently, numerous face spoofing detection schemes have been proposed based on the assumption that training and testing samples are in the same domain in terms of the feature space and marginal probability distribution. However, due to unlimited variations of the dominant conditions (illumination, facial appearance, camera quality, and so on) in face acquisition, such single domain methods lack generalization capability, which further prevents them from being applied in practical applications. In light of this, we introduce an unsupervised domain adaptation face anti-spoofing scheme to address the real-world scenario that learns the classifier for the target domain based on training samples in a different source domain. In particular, an embedding function is first imposed based on source and target domain data, which maps the data to a new space where the distribution similarity can be measured. Subsequently, the Maximum Mean Discrepancy between the latent features in source and target domains is minimized such that a more generalized classifier can be learned. State-of-the-art representations including both hand-crafted and deep neural network learned features are further adopted into the framework to quest the capability of them in domain adaptation. Moreover, we introduce a new database for face spoofing detection, which contains more than 4000 face samples with a large variety of spoofing types, capture devices, illuminations, and so on. Extensive experiments on existing benchmark databases and the new database verify that the proposed approach can gain significantly better generalization capability in cross-domain scenarios by providing consistently better anti-spoofing performance.

Unsupervised Domain Adaptation for Face Anti-Spoofing

In this paper, we propose a novel framework leveraging the advantages of the representational ability of deep learning and domain generalization for face spoofing detection. In particular, the generalized deep feature representation is achieved by taking both spatial and temporal information into consideration, and a 3D convolutional neural network architecture tailored for the spatial-temporal input is proposed. The network is first initialized by training with augmented facial samples based on cross-entropy loss and further enhanced with a specifically designed generalization loss, which coherently serves as the regularization term. The training samples from different domains can seamlessly work together for learning the generalized feature representation by manipulating their feature distribution distances. We evaluate the proposed framework with different experimental setups using various databases. Experimental results indicate that our method can learn more discriminative and generalized information compared with the state-of-the-art methods.

Haoliang Li

Papers

Domain Generalization with Adversarial Feature Learning

Unsupervised Domain Adaptation for Face Anti-Spoofing

Learning Generalized Deep Feature Representation for Face Anti-Spoofing

Experimental determination of dynamic tensile properties of a granite

Dynamic uniaxial compression tests on a granite