We have collection of more than 1 Million open source products ranging from Enterprise product to
small libraries in all platforms. We aggregate information from all open source repositories.
Search and find the best for your needs. Check out projects section.
Image-to-Image Translation with Conditional Adversarial Networks Phillip Isola, Jun-Yan Zhu, Tinghui Zhou, Alexei A. Efros CVPR, 2017. On some tasks, decent results can be obtained fairly quickly and on small datasets. For example, to learn to generate facades (example shown above), we trained on just 400 images for about 2 hours (on a single Pascal Titan X GPU). However, for harder problems it may be important to train on far larger datasets, and for many hours or even days.
[Project] [Youtube] [Paper] A research prototype developed by UC Berkeley and Adobe CTL. Latest development: [pix2pix]: Torch implementation for learning a mapping from input images to output images. [CycleGAN]: Torch implementation for learning an image-to-image translation (i.e. pix2pix) without input-output pairs. [pytorch-CycleGAN-and-pix2pix]: PyTorch implementation for both unpaired and paired image-to-image translation.
All have been tested with python2.7+ and tensorflow1.0+ in linux. The final layer can be sigmoid(data: [0,1]) or tanh(data:[-1,1]), my codes all use sigmoid. Using weights_initializer=tf.random_normal_initializer(0, 0.02) will converge faster.
This repository collects chainer implementation of state-of-the-art GAN algorithms. These codes are evaluated with the inception score on Cifar-10 dataset. Note that our codes are not faithful re-implementation of the original paper. This implementation has been tested with the following versions.
This is a short implementation of a Conditional DCGAN, however if you need a cDCGAN for real-world use cases, please consider using a more serious implementation. Here can be seen a cDCGAN trained on CIFAR-10 using the same networks architectures I used for MNIST, obviously it shows that we need to be careful when designing the architecture. It works better using more filters.
This repository contains code to instantiate and deploy an image completer model. The goal is to fill in missing or corrupted parts of an image. This model uses Deep Convolutional Generative Adversarial Networks (DCGAN) to fill the missing regions in an image. The model is trained on the celebA dataset and works best for completing corrupted portions of a human face. Input to the model is an image containing a single corrupted face. The OpenFace face recognition tool will detect and extract the corrupted face from the input image. This extracted face is then passed to the OpenFace alignment tool where it is aligned (inner eyes with bottom lip) and resized (64 x 64) producing an output that can be used by the model to fill the corrupted portions. The output is a collage of 20 images, in a 4x5 grid, representing the intermediate results and final completed image (bottom-right). The model is based on the Tensorflow implementation of DCGAN. The model weights are hosted on IBM Cloud Object Storage. The code in this repository deploys the model as a web service in a Docker container. This repository was developed as part of the IBM Developer Model Asset Exchange.
Implementation of DCGAN with TensorFlow slim. Base codes and models are from DCGAN in Tensorflow made by Taehoon Kim. At this time, this code only support Flower dataset, but maybe with some tweaks you can train/evaluate in other dataset. I know there are lots of code of DCGAN, especially made by Taehoon Kim. However, this code implement DCGAN with the bleeding edges features of TensorFlow such as TF-Slim, tf.train.Supervisor and TFRecords etc.