Nvidia to Open StyleGan Source Code

This machine learning project allows to create of people faces by imitating photographs
11 February 2019   775

NVIDIA has open source code if developments related to the StyleGAN project, which allows generating images of new faces of people by imitating photographs. The system automatically takes into account aspects of the placement of individuals and makes the result indistinguishable from real photos (most of the respondents could not distinguish the original photos from the generated ones). For the synthesis of individuals, a machine learning system based on a generative-competitive neural network (GAN) is used. The code is written in Python using the TensorFlow framework and published under the Creative Commons BY-NC 4.0 license (for non-commercial use only).

Both ready-made trained models and collections of images for self-learning of a neural network are available for download. The basic model was trained on the basis of the Flickr-Faces-HQ (FFHQ) collection, which includes 70,000 high-quality (1024x1024) PNG images of people's faces. At the same time, the system is not tied to persons - as an example, the variants trained on collections of photographs of cars, cats and beds are shown. It requires one or more NVIDIA graphics cards (Tesla V100 GPU recommended), at least 11 GB of RAM, NVIDIA 391.35+ drivers, CUDA 9.0+ tools and the cuDNN 7.3.1 library.

The system allows you to synthesize the image of a new face based on interpolation of features of several faces, combining features characteristic of them, as well as adapting the final image to the required age, gender, hair length, smile character, nose shape, skin color, glasses, face rotation in the photo. The generator considers the image as a collection of styles, automatically separates the characteristic details (freckles, hair, glasses) from common high-level attributes (posture, gender, age changes) and allows you to combine them in an arbitrary form with the definition of the dominant properties through weights.

Python to be Language of 2018 According to TIOBE

According to experts, Python has become an integral part of many IT-spheres, including AI
09 January 2019   433

TIOBE experts called Python the programming language of 2018 - according to them, Python has become an integral part of many IT-spheres. It is leader of use at statistics, artificial intelligence systems development, scripts and system tests, and is still widely used in web development and scientific computing.

By the way, Python reached this peak for the third time - none of the languages ​​showed this result.

TIOBE Languages of the Year List
TIOBE Languages of the Year List

TIOBE experts summed up not only the whole year, but also the month. Returning to the top three in December, in January, Python continued to push C ++, although the last time it lost ground fairly quickly.

MATLAB changed places with Ruby: the first one rose to the 11th line, the second one dropped to 18. TypeScript took off as much as 118 points and took the 49th place. F # lost two dozen points, and Alice lost all four.

Tiobe Index Jan 2019 to Jan 2018 Compare
Tiobe Index Jan 2019 to Jan 2018 Compare

Kotlin started the year with a positive. The language has risen to 31st place, and experts believe that in 2019 it will enter the top 20.