Autoencoder-learned local image descriptor for image inpainting
In this paper, we propose an efficient method for learning local image descriptors suitable for the use in image inpainting algorithms. We learn the descriptors using a convolutional autoencoder network that we design such that the network produces a computationally efficient extraction of patch descriptors through an intermediate image representation. We integrate our descriptor into an inpainting algorithm to show computational memory and time benefits of the employment of our descriptor.
I am Nina Žižakić, a PhD student at Group for Artificial Intelligence and Sparse Modelling (GAIM) in Ghent University, working with prof. Aleksandra Pižurica. My interests are in the fields of artificial intelligence and image processing. I am currently working on learning local feature descriptors with the use of autoencoders.