arXiv subjects

Attila Szabó

Publications and source records attributed to Attila Szabó.

22 records · Page 2Linked to original sources

arXiv2018

Non-power-law universality in one-dimensional quasicrystals

We have investigated scaling properties of the Aubry-André model and related one-dimensional quasiperiodic Hamiltonians near their localisation transitions. We find numerically that the scaling of characteristic energies near the ground state, usually captured by a single dynamical exponent, does not obey a power law relation. Instead, the scaling behaviour depends strongly on the correlation length in a manner governed by the continued fraction expansion of the irrational number $β$ describing incommensurability in the system. This dependence is, however, found to be universal between a range of models sharing the same value of $β$. For the Aubry-André model, we explain this behaviour in terms of a discrete renormalisation group protocol which predicts rich critical behaviour. This result is complemented by studies of the expansion dynamics of a wave packet under the Aubry-André model at the critical point. Anomalous diffusion exponents are derived in terms of multifractal (Rényi) dimensions of the critical spectrum; non-power-law universality similar to that found in ground state dynamics is observed between a range of critical tight-binding Hamiltonians.

cond-mat.dis-nn↗

arXiv2018

FaceShop: Deep Sketch-based Face Image Editing

We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.

cs.CV↗

arXiv2018

Disentangling Factors of Variation by Mixing Them

We propose an approach to learn image representations that consist of disentangled factors of variation without exploiting any manual labeling or data domain knowledge. A factor of variation corresponds to an image attribute that can be discerned consistently across a set of images, such as the pose or color of objects. Our disentangled representation consists of a concatenation of feature chunks, each chunk representing a factor of variation. It supports applications such as transferring attributes from one image to another, by simply mixing and unmixing feature chunks, and classification or retrieval based on one or several attributes, by considering a user-specified subset of feature chunks. We learn our representation without any labeling or knowledge of the data domain, using an autoencoder architecture with two novel training objectives: first, we propose an invariance objective to encourage that encoding of each attribute, and decoding of each chunk, are invariant to changes in other attributes and chunks, respectively; second, we include a classification objective, which ensures that each chunk corresponds to a consistently discernible attribute in the represented image, hence avoiding degenerate feature mappings where some chunks are completely ignored. We demonstrate the effectiveness of our approach on the MNIST, Sprites, and CelebA datasets.

cs.CV↗

arXiv2017

Challenges in Disentangling Independent Factors of Variation

We study the problem of building models that disentangle independent factors of variation. Such models could be used to encode features that can efficiently be used for classification and to transfer attributes between different images in image synthesis. As data we use a weakly labeled training set. Our weak labels indicate what single factor has changed between two data samples, although the relative value of the change is unknown. This labeling is of particular interest as it may be readily available without annotation costs. To make use of weak labels we introduce an autoencoder model and train it through constraints on image pairs and triplets. We formally prove that without additional knowledge there is no guarantee that two images with the same factor of variation will be mapped to the same feature. We call this issue the reference ambiguity. Moreover, we show the role of the feature dimensionality and adversarial training. We demonstrate experimentally that the proposed model can successfully transfer attributes on several datasets, but show also cases when the reference ambiguity occurs.

cs.CV↗