Open AccessProceedings Article
Arbitrary Style Transfer in Real-time with Adaptive Instance Normalization.
Xun Huang,Serge Belongie +1 more
Reads0
Chats0
TLDR
In this article, adaptive instance normalization (AdaIN) is proposed to align the mean and variance of the content features with those of the style features, which enables arbitrary style transfer in real-time.Abstract:
Gatys et al. recently introduced a neural algorithm that renders a content image in the style of another image, achieving so-called style transfer. However, their framework requires a slow iterative optimization process, which limits its practical application. Fast approximations with feed-forward neural networks have been proposed to speed up neural style transfer. Unfortunately, the speed improvement comes at a cost: the network is usually tied to a fixed set of styles and cannot adapt to arbitrary new styles. In this paper, we present a simple yet effective approach that for the first time enables arbitrary style transfer in real-time. At the heart of our method is a novel adaptive instance normalization (AdaIN) layer that aligns the mean and variance of the content features with those of the style features. Our method achieves speed comparable to the fastest existing approach, without the restriction to a pre-defined set of styles. In addition, our approach allows flexible user controls such as content-style trade-off, style interpolation, color & spatial controls, all using a single feed-forward neural network.read more
Citations
More filters
Proceedings Article
Segment2Regress: Monocular 3D Vehicle Localization in Two Stages.
TL;DR: This paper proposes a monocular camera-based 3D vehicle localization method that alleviates the dependency on high-quality depth maps by taking advantage of the commonly accepted assumption that the observed vehicles lie on the road surface, called Segment2Regress.
BookDOI
Multimedia Communications, Services and Security : 10th International Conference, MCSS 2020, Kraków, Poland, October 8-9, 2020, Proceedings
TL;DR: This paper presents a new approach to analyse cybersecurity challenges and opportunities, focused on the development of a new risk assessment and management framework that is applicable in healthcare, energy, maritime transportation, and defence sectors, being extensible to others.
Journal ArticleDOI
Synthesizing data for text recognition with style transfer
TL;DR: This work proposes a word data generating method called SynthText-Transfer, which is capable of emulating the distribution of the target dataset, and uses a style transfer method to generate samples with arbitray text content, which preserve the texture of the reference sample in the target datasets.
Posted Content
Conceptual Compression via Deep Structure and Texture Synthesis.
TL;DR: A novel conceptual compression framework that encodes visual data into compact structure and texture representations, then decodes in a deep synthesis fashion, aiming to achieve better visual reconstruction quality, flexible content manipulation, and potential support for various vision tasks is proposed.
Book ChapterDOI
Generative Creativity: Adversarial Learning for Bionic Design
TL;DR: This work proposes DesignGAN, a novel unsupervised deep generative approach to realising bionic design that employs a conditional Generative Adversarial Networks architecture with several designated losses that respectively constrict the model to meet the corresponding requirements of bionics design modelling.
References
More filters
Proceedings Article
Adam: A Method for Stochastic Optimization
Diederik P. Kingma,Jimmy Ba +1 more
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
Journal ArticleDOI
Generative Adversarial Nets
Ian Goodfellow,Jean Pouget-Abadie,Mehdi Mirza,Bing Xu,David Warde-Farley,Sherjil Ozair,Aaron Courville,Yoshua Bengio +7 more
TL;DR: A new framework for estimating generative models via an adversarial process, in which two models are simultaneously train: a generative model G that captures the data distribution and a discriminative model D that estimates the probability that a sample came from the training data rather than G.
Proceedings Article
Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
Sergey Ioffe,Christian Szegedy +1 more
TL;DR: Applied to a state-of-the-art image classification model, Batch Normalization achieves the same accuracy with 14 times fewer training steps, and beats the original model by a significant margin.
Book ChapterDOI
Microsoft COCO: Common Objects in Context
Tsung-Yi Lin,Michael Maire,Serge Belongie,James Hays,Pietro Perona,Deva Ramanan,Piotr Dollár,C. Lawrence Zitnick +7 more
TL;DR: A new dataset with the goal of advancing the state-of-the-art in object recognition by placing the question of object recognition in the context of the broader question of scene understanding by gathering images of complex everyday scenes containing common objects in their natural context.