In this paper we propose Spatial PixelCNN, a conditional autoregressive model that generates images from small patches. By conditioning on a grid of pixel coordinates and global features extracted from a Variational Autoencoder (VAE), we are able to train on patches of images, and reproduce the full-sized image. We show that it not only allows for generating high quality samples at the same res...

The network-based machine learning algorithm is very powerful tools. However, it requires huge training dataset. Researchers often meet privacy issues when they collect image dataset especially for surveillance applications. A learnable image encryption scheme is introduced. The key idea of this scheme is to encrypt images, so that human cannot understand images but the network can be train wit...

Although Deep Convolutional Neural Networks trained with strong pixel-level annotations have significantly pushed the performance in semantic segmentation, annotation efforts required for the creation of training data remains a roadblock for further improvements. We show that augmentation of the weakly annotated training dataset with synthetic images minimizes both the annotation efforts and al...

Introduction: Nowadays, magnetic resonance imaging (MRI) in combination with computed-tomography (CT) is increasingly being used in radiation therapy planning. MR and CT images are applied to determine the target volume and calculate dose distribution, respectively. Since the use of these two imaging modalities causes registration uncertainty and increases department w...

On the research work leading to automatic detection of optic disc from retinal images is very essential and crucial for expert ophthalmologists to diagnose diseases. Many of techniques can achieve good performance on retinal feature that is clearly visible. Unfortunately, it is a normal situation that the color retinal images in Thailand are poor-quality images. The existing algorithm cannot de...

A. Lanzara, 2 P. V. Bogdanov, X. J. Zhou, N. Kaneko, H. Eisaki, M. Greven, Z. Hussain, and Z. -X. Shen Department of Physics, University of California, Berkeley, California 94720, USA Materials Sciences Division, Lawrence Berkley National Laboratory, Berkeley, California 94720, USA Department of Physics, Applied Physics and Stanford Synchrotron Radiation Laboratory, Stanford University, Stanfor...

Many modern and popular state of the art image denoising algorithms are trained and evaluated using images corrupted by artificial noise. These trained algorithms and their evaluations on synthetic data may lead to incorrect conclusions about their performances on real noise. In this paper we introduce a benchmark dataset of uncompressed color images corrupted by natural noise due to low-light ...

background: medical image interpolation is recently introduced as a helpful tool to obtain further information via initial available images taken by tomography systems. this information may be useful for better diagnosis of possible lesions or better tumor delineation at radiation treatment. to do this, deformable image registration algorithms are mainly utilized to perform image interpolation ...

The goal of this work is to retrieve images containing both a target person and a target scene type from a large dataset of images. At run time this compound query is handled using a face classifier trained for the person, and an image classifier trained for the scene type. We make three contributions: first, we propose a hybrid convolutional neural network architecture that produces place-desc...