FAIR Released Detectron2 – New Object Detection Library

14 October 2019

FAIR Released Detectron2 – New Object Detection Library

Facebook AI Research (FAIR) has announced the release of Detectron2 – a PyTorch-based object detection library as the second version of Detectron, released last year. The new library is built…

AdaptIS: New Instance Segmentation Method From Samsung AI

11 October 2019

AdaptIS: New Instance Segmentation Method From Samsung AI

Researchers from Samsung AI Center have proposed a novel method for instance segmentation that achieves state-of-the-art results on several benchmark datasets. AdaptIS or Adaptive Instance Selection Network is a class-agnostic…

DeepMind Researchers Trained RL Agents To Paint Realistic Portraits

7 October 2019

DeepMind Researchers Trained RL Agents To Paint Realistic Portraits

A group of researchers from Google DeepMind has published their recent work at the intersection of deep reinforcement learning (Deep RL) and Generative Adversarial Networks training. In their paper named…

Neural Network For Image Super-resolution Trained in Completely Unsupervised Manner

4 October 2019

Neural Network For Image Super-resolution Trained in Completely Unsupervised Manner

Researchers from ETH Zurich’s Computer Vision Lab have proposed a new, fully unsupervised method for image super-resolution. According to the evaluation, the novel method outperforms current state-of-the-art methods in this…

Facebook Uses Deep Neural Networks to Track Hand Movements in Oculus

30 September 2019

Facebook Uses Deep Neural Networks to Track Hand Movements in Oculus

Engineers from Facebook Reality Labs and Oculus have developed a fully articulated hand tracking system for Virtual Reality (VR) that is based on completely on monochrome cameras. The new system…

AI Agents Learnt How To Use Tools And Exploit The Environment in a Hide-and-Seek Game

21 September 2019

AI Agents Learnt How To Use Tools And Exploit The Environment in a Hide-and-Seek Game

The latest research from OpenAI researchers shows that AI agents are able to use clever tricks and manipulate the environment to reach their predefined goal. In the novel paper “Emergent…

Google’s AI Can Accurately Detect 26 Skin Deseases

20 September 2019

Google’s AI Can Accurately Detect 26 Skin Deseases

Researchers from Google, the University of California, MIT and the Medical University of Graz have joined forces to develop a deep learning system for diagnosis of a large number of…

DeepPrivacy: GAN Network Anonymizes Images Using Generated Faces

14 September 2019

DeepPrivacy: GAN Network Anonymizes Images Using Generated Faces

A group of researchers from the Norwegian University of Science and Technology has proposed a method for automatic face anonymization by superimposing a generated face on the input image. The…

SynVAE: Neural Network That Translates Visual Art Into Music

13 September 2019

SynVAE: Neural Network That Translates Visual Art Into Music

Researchers from the University of Amsterdam proposed a method that can translate visual art into music. Their method, called SynVAE or Synesthetic Variational Autoencoder leverages the power of deep autoencoder…

DeepMind Released Four New Multi-Object Datasets

7 September 2019

DeepMind Released Four New Multi-Object Datasets

Deepmind researchers have released new datasets for multi-object representation learning. The datasets (4 in total) contain multi-object scene images, where each of the images is accompanied by ground truth data…

Neural Network Generates In-Game Characters of Players From A Single Photo

4 September 2019

Neural Network Generates In-Game Characters of Players From A Single Photo

A group of researchers from NetEase Fuxi AI Lab has developed a method that uses deep learning to generate in-game characters of players according to an input image. The proposed…

Chinese Deepfake App Zao Uses AI To Turn You Into a Hollywood Actor

3 September 2019

Chinese Deepfake App Zao Uses AI To Turn You Into a Hollywood Actor

A new Chinese app that uses deepfakes AI has become one of the most downloaded mobile applications after its launch on Friday in China. The app that turns a selfie…

Are Weights Really Important? Researchers From Google Proposed Weight-Agnostic Neural Networks

3 September 2019

Are Weights Really Important? Researchers From Google Proposed Weight-Agnostic Neural Networks

Researchers from Google have proposed a new type of deep neural networks – Weight-agnostic Neural Networks, which are able to perform various tasks even when they use random weights. In…

Neural Network Can Generate Up To 8x Super-resolved Images

31 August 2019

Neural Network Can Generate Up To 8x Super-resolved Images

Researchers from the Korean Advanced Institute of Science and Technology, have developed a method for face Super-Resolution that can generate photo-realistic 8x super-resolved images. The new method is based on…

UAR: New Metric For Testing Robustness Against Unforeseen Adversarial Attacks

24 August 2019

UAR: New Metric For Testing Robustness Against Unforeseen Adversarial Attacks

In their new paper, researchers from OpenAI introduce a new metric which evaluates the robustness of neural network classifiers against unforeseen adversarial examples. The new metric, called UAR (Unforeseen Attack…

OpenAI Released The Largest GPT-2 Model With 774 Million Parameters

21 August 2019

OpenAI Released The Largest GPT-2 Model With 774 Million Parameters

Six months after OpenAI published the paper on their new, large and powerful generative model for NLP – GPT-2, today the non-profit announced that it is releasing the 774 million…

Google Released New Real-time Hand Tracking Algorithm

21 August 2019

Google Released New Real-time Hand Tracking Algorithm

Valentin Bazarevsky and Fan Zhang, Research Engineers from Google Research have presented a real-time hand tracking software that can run on mobile devices. Deep learning-based methods have shown unprecedented performance…

U-GAT-IT – New State-of-the-art Method For Image-to-Image Translation

17 August 2019

U-GAT-IT – New State-of-the-art Method For Image-to-Image Translation

A group of researchers from NCSOFT and Boeing Korea Engineering and Technology Center has proposed a novel state-of-the-art method for image-to-image translation. By introducing several novelties, researchers managed to build…

DeepMind Uses Machine Learning To Track Serengeti Wildlife

15 August 2019

DeepMind Uses Machine Learning To Track Serengeti Wildlife

Researchers from Deepmind have joined forces with ecologists and conservationists to use machine learning for the purpose of studying wildlife behavior in Tanzania’s Serengeti National Park. The project builds upon…

Google AI Released A Novel Self-Supervised Method for Video Understanding

13 August 2019

Google AI Released A Novel Self-Supervised Method for Video Understanding

A group of researchers from Google Research and Deepmind has proposed a novel, self-supervised method for video understanding. The new method, called Temporal Cycle-Consistency Learning (TCC) is able to align…

Researchers Used Machine Learning To Map Entire Fruit Fly Brain

6 August 2019

Researchers Used Machine Learning To Map Entire Fruit Fly Brain

Researchers from Google, in collaboration with researchers from Janelia Research Campus and Cambridge University, have presented a method for fully automated reconstruction of an entire fruit fly brain. In their…