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

High-resolution representations are essential for position-sensitive vision problems, such as human pose estimation, semantic segmentation, and object detection. Existing state-of-the-art frameworks first encode the input image as a low-resolution representation through a subnetwork that is formed by connecting high-to-low resolution convolutions \emph{in series} (e.g., ResNet, VGGNet), and then recover the high-resolution representation from the encoded low-resolution representation. Instead, our proposed network, named as High-Resolution Network (HRNet), maintains high-resolution representations through the whole process. There are two key characteristics: (i) Connect the high-to-low resolution convolution streams \emph{in parallel}; (ii) Repeatedly exchange the information across resolutions. The benefit is that the resulting representation is semantically richer and spatially more precise. We show the superiority of the proposed HRNet in a wide range of applications, including human pose estimation, semantic segmentation, and object detection, suggesting that the HRNet is a stronger backbone for computer vision problems. All the codes are available at~{\url{this https URL}}.

Deep High-Resolution Representation Learning for Visual Recognition

High-resolution representations are essential for position-sensitive vision problems, such as human pose estimation, semantic segmentation, and object detection. Existing state-of-the-art frameworks first encode the input image as a low-resolution representation through a subnetwork that is formed by connecting high-to-low resolution convolutions in series (e.g., ResNet, VGGNet), and then recover the high-resolution representation from the encoded low-resolution representation. Instead, our proposed network, named as High-Resolution Network (HRNet), maintains high-resolution representations through the whole process. There are two key characteristics: (i) Connect the high-to-low resolution convolution streams in parallel and (ii) repeatedly exchange the information across resolutions. The benefit is that the resulting representation is semantically richer and spatially more precise. We show the superiority of the proposed HRNet in a wide range of applications, including human pose estimation, semantic segmentation, and object detection, suggesting that the HRNet is a stronger backbone for computer vision problems. All the codes are available at https://github.com/HRNet .

/pdf/deep-high-resolution-representation-learning-for-visual-9n1rjbo57m.pdf

The detection of anomalous structures in natural image data is of utmost importance for numerous tasks in the ﬁeld of computer vision. The development of methods for unsupervised anomaly detection requires data on which to train and evaluate new approaches and ideas. We introduce the MVTec Anomaly Detection (MVTec AD) dataset containing 5354 high-resolution color images of different object and texture categories. It contains normal, i.e., defect-free, images intended for training and images with anomalies intended for testing. The anomalies manifest themselves in the form of over 70 different types of defects such as scratches, dents, contaminations, and various structural changes. In addition, we provide pixel-precise ground truth regions for all anomalies. We also conduct a thorough evaluation of current state-of-the-art unsupervised anomaly detection methods based on deep architectures such as convolutional autoencoders, generative adversarial networks, and feature descriptors using pre-trained convolutional neural networks, as well as classical computer vision methods. This initial benchmark indicates that there is considerable room for improvement. To the best of our knowledge, this is the ﬁrst comprehensive, multi-object, multi-defect dataset for anomaly detection that provides pixel-accurate ground truth regions and focuses on real-world applications.

/pdf/mvtec-ad-a-comprehensive-real-world-dataset-for-unsupervised-3029ppuqb1.pdf

MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection

Anomaly detection is an important problem that has been well-studied within diverse research areas and application domains. The aim of this survey is two-fold, firstly we present a structured and comprehensive overview of research methods in deep learning-based anomaly detection. Furthermore, we review the adoption of these methods for anomaly across various application domains and assess their effectiveness. We have grouped state-of-the-art research techniques into different categories based on the underlying assumptions and approach adopted. Within each category we outline the basic anomaly detection technique, along with its variants and present key assumptions, to differentiate between normal and anomalous behavior. For each category, we present we also present the advantages and limitations and discuss the computational complexity of the techniques in real application domains. Finally, we outline open issues in research and challenges faced while adopting these techniques.

Deep Learning for Anomaly Detection: A Survey.

In this work we propose HR-Dehazer, a novel and accurate method for image dehazing. An encoder-decoder neural network is trained to learn a direct mapping between a hazy image and its respective clear version. We designed a special loss that forces the network to keep into account the semantics of the input image and to promote consistency among local structures. In addition, this loss makes the system more invariant to scale changes. Quantitative results on the recently released DenseHaze dataset introduced for the NTIRE2019-Dehazing challenge demonstrates the effectiveness of the proposed method. Furthermore, qualitative results on real data show that the described solution generalizes well to different never-seen scenarios.

/pdf/high-resolution-single-image-dehazing-using-encoder-decoder-1xnfr64iu6.pdf

High-Resolution Single Image Dehazing Using Encoder-Decoder Architecture

In this work we present SpliNet, a novel CNN-based method that estimates a global color transform for the enhancement of raw images. The method is designed to improve the perceived quality of the images by reproducing the ability of an expert in the field of photo editing. The transformation applied to the input image is found by a convolutional neural network specifically trained for this purpose. More precisely, the network takes as input a raw image and produces as output one set of control points for each of the three color channels. Then, the control points are interpolated with natural cubic splines and the resulting functions are globally applied to the values of the input pixels to produce the output image. Experimental results compare favorably against recent methods in the state of the art on the MIT-Adobe FiveK dataset. Furthermore, we also propose an extension of the SpliNet in which a single neural network is used to model the style of multiple reference retouchers by embedding them into a user space. The style of new users can be reproduced without retraining the network, after a quick modeling stage in which they are positioned in the user space on the basis of their preferences on a very small set of retouched images.

Personalized Image Enhancement Using Neural Spline Color Transforms

In this work we propose a novel CNN-based method for image enhancement that simulates an expert retoucher. The method is fast and accurate at the same time thanks to the decoupling between the inference of the parameters and the color transformation. Specifically, the parameters are inferred from a downsampled version of the raw input image and the transformation is applied to the full resolution input. Different variants of the proposed enhancement method can be generated by varying the parametric functions used as color transformations (i.e. polynomial, piecewise, cosine and radial), and by varying how they are applied (i.e. channelwise or full color). Experimental results show that several variants of the proposed method outperform the state of the art on the MIT-Adobe FiveK dataset.

Learning Parametric Functions for Color Image Enhancement

We propose a novel method based on Convolutional Neural Networks for content-preserving tone adjustment. The method is at the same time fast and accurate since we decouple the inference of the parameters and the color transform: the parameters are inferred from a downsampled version of the input image and the transformation is applied to the full resolution input. The method includes two steps of image enhancement: the first one is a global color transformation, while the second one is a local transformation. Experiments conducted on the DPED - DSLR Photo Enhancement Dataset, that has been used for the NTIRE19 Image Enhancement Challenge, and on the MIT-Adobe FiveK dataset, that is widely used for image enhancement, demonstrate the effectiveness of the proposed method.

/pdf/content-preserving-tone-adjustment-for-image-enhancement-2h6cx3hawe.pdf

Content-Preserving Tone Adjustment for Image Enhancement

/pdf/unsupervised-segmentation-of-hyperspectral-remote-sensing-1vi1beyd.pdf

Flavio Piccoli

Papers

High-Resolution Single Image Dehazing Using Encoder-Decoder Architecture

Personalized Image Enhancement Using Neural Spline Color Transforms

Learning Parametric Functions for Color Image Enhancement

Content-Preserving Tone Adjustment for Image Enhancement

Unsupervised Segmentation of Hyperspectral Remote Sensing Images with Superpixels