Patent
Cyclic generative adversarial network for unsupervised cross-domain image generation
Choi Wongun,Schulter Samuel,Sohn Kihyuk,Chandraker Manmohan +3 more
- 25 Oct 2018
17
TL;DR: In this paper, a cross-domain image generation system is proposed for unsupervised image generation relative to a first and second image domain that each include real images, where a first generator generates synthetic images similar to real images in the second domain while including a semantic content of real image in the first domain.
read more
Abstract: A system is provided for unsupervised cross-domain image generation relative to a first and second image domain that each include real images. A first generator generates synthetic images similar to real images in the second domain while including a semantic content of real images in the first domain. A second generator generates synthetic images similar to real images in the first domain while including a semantic content of real images in the second domain. A first discriminator discriminates real images in the first domain against synthetic images generated by the second generator. A second discriminator discriminates real images in the second domain against synthetic images generated by the first generator. The discriminators and generators are deep neural networks and respectively form a generative network and a discriminative network in a cyclic GAN framework configured to increase an error rate of the discriminative network to improve synthetic image quality.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Method for learning cross-domain relations based on generative adversarial networks
Kim Taek Soo,Cha Moon Su,Kim Ji Won +2 more
- 01 Nov 2018
TL;DR: In this article, a generative adversarial networks-based method for cross-domain relations is proposed, which includes two coupled GANs: a first GAN learns a translation of images from domain A to domain B, and a second GAN learning a translation from domain B to domain A.
31
Patent
Method for image analysis
Carl Magnus Wrenninge,Carl Jonas Magnus Unger +1 more
- 23 May 2019
TL;DR: In this article, a method for synthetic data generation and analysis including determining a set of parameter values, generating a scene based on the parameter values and rendering a synthetic image of the scene was proposed.
14
Patent
Synthesizing and segmenting cross-domain medical images
Zheng Yefeng,Zhang Zizhao +1 more
- 28 Feb 2019
TL;DR: In this paper, the first generator is trained based on a comparison between segmentation results of a training image in the first domain from a first segmentor and segmentation result of a synthesized training image from a second segmentor.
8
Patent
Label box-free micro-seismic signal detection method and device
Sheng Guanqun,Yang Chao,Tang Xingong,Kai Xie,Tang Jing +4 more
- 02 Aug 2019
TL;DR: In this article, a label box-free micro-seismic signal detection method and device was proposed, which comprises the following steps: two data sets a and b containing effective signals are screened out from micro-SEISMic signal data, and preprocessing is carried out through a convolutional neural network; by means of an RPN layer in a Faster-RCNN, generating candidate boxes on data feature graphs; discriminating the similarity of the candidate boxes by a discriminator of a generative adversarial network, thereby obtaining the candidate box with the similarity exceeding a preset
4
Patent
method for generating a deposit training sample by using a generative adversarial network
Zhou Min,Zhu Zhichao,Wang Yong,Yang Jian,Zeng Yuan,Tuersunaili +5 more
- 11 Jun 2019
TL;DR: In this paper, a method for generating a deposit training sample by using a generative adversarial network was proposed, which comprises the following steps of creating a one-to-one correspondence training set of deposits and points; using a discriminator and a generator to train a GAN model; enabling the random generator to generate a random distribution set of the points of the plane; generating more samples by using the generated random points; and optimizing the GAN models by using generated sample.
2
References
Perceptual Losses for Real-Time Style Transfer and Super-Resolution
Justin Johnson,Alexandre Alahi,Li Fei-Fei +2 more
- 08 Oct 2016
TL;DR: In this paper, the authors combine the benefits of both approaches, and propose the use of perceptual loss functions for training feed-forward networks for image style transfer, where a feedforward network is trained to solve the optimization problem proposed by Gatys et al. in real-time.
•Posted Content
Learning to Discover Cross-Domain Relations with Generative Adversarial Networks
TL;DR: This work proposes a method based on generative adversarial networks that learns to discover relations between different domains (DiscoGAN) and successfully transfers style from one domain to another while preserving key attributes such as orientation and face identity.
2K
Patent
System and method for image annotation and multi-modal image retrieval using probabilistic semantic models
Ruofei Zhang,Zhongfei Zhang +1 more
- 24 Jan 2007
TL;DR: In this article, an Expectation-Maximization (EM) based iterative learning procedure determines the conditional probabilities of the visual features and the textual words given a hidden concept class.
22
Patent
Systems and methods for image-to-image translation using variational autoencoders
Ming-Yu Liu,Thomas M. Breuel,Jan Kautz +2 more
- 30 Aug 2018
TL;DR: In this paper, a method, computer readable medium, and system are disclosed for training a neural network, which includes the steps of encoding, by a first neural network and a second neural network.
14
Patent
Estimator training method and pose estimating method using depth image
Jae Joon Han,Danhang Tang,Tae-Kyun Kim,Han Seungju,Yoo Byung In,Chang Kyu Choi,Alykhan Tejani,Hyung Jin Chang +7 more
- 19 May 2014
TL;DR: In this article, an estimator training method and a pose estimation method using a depth image are disclosed, in which the estimation method may train an estimators configured to estimate a pose of an object, based on an association between synthetic data and real data, and the pose estimating method may estimate the pose of the object using the trained estimators.
10