https://bit.ly/2KQTfzF

Face Recognition is the upcoming and modern challenge required in the machine learning in today's world. We have listed out some of the best Face Recogntion APIs which can be seamlessly integrated into your project to go one step further in image processing

AI Habitat: an advanced simulation platform for embodied AI research
Written by Manolis Savva, Abhishek Kadian, Oleksandr Maksymets, Dhruv Batra: https://ai.facebook.com/blog/open-sourcing-ai-habitat-an-simulation-platform-for-embodied-ai-research/
#ArtificialIntelligence #DeepLearning #ReinforcementLearning

Cool paper from Microsoft Research team achieving SOTA provable L2-robustness on ImageNet by adversarially training a neural network convolved with Gaussian noise!

paper: https://arxiv.org/abs/1906.04584
code: https://github.com/Hadisalman/smoothing-adversarial
blog: https://decentdescent.org/smoothadv.html

I am excited to share work from my team at Facebook Reality Labs: the Replica Dataset - a high quality dataset of 18 3D reconstruction that has clean dense geometry, high resolution and high dynamic range textures, glass and mirror surface information, and semantic class and instance segmentation. See https://arxiv.org/abs/1906.05797 for more details. You can download Replica v1 now via https://github.com/facebookresearch/Replica-Dataset

This was a joint effort with FAIR and their awesome AI Habitat Simulator (https://aihabitat.org/). Here are two blog posts describing how Replica and AI Habtiat fit together to train the next generation of AI agents and assistants:
https://tech.fb.com/facebook-reality-labs-replica-simulations-help-advance-ai-and-ar/
https://ai.facebook.com/blog/open-sourcing-ai-habitat-an-simulation-platform-for-embodied-ai-research

Yann lecun :

FAIR and Facebook Reality Lab (FRL) have collaborated to release two interactive environments for trainin AI agents:
1. AI Habitat: fast simulation of indoor environments
2. Replica: visually realistic indoor environment
Blog posts:
https://ai.facebook.com/blog/open-sourcing-ai-habitat-an-simulation-platform-for-embodied-ai-research/
https://tech.fb.com/facebook-reality-labs-replica-simulations-help-advance-ai-and-ar/

SLIDES
Generating high Fidelity Images with Subscale Pixel Networks and Multidimensional Upscaling

Jacob Menick Nal Kalchbrenner
DeepMind Google Brain Amsterdam

https://drive.google.com/file/d/1bbJrQmCAjzkEZpumWQClo_qR3wBQFWD8/view?fbclid=IwAR2Z2UZAfqiw6o-2ctpCAOj8njzHnHc-sSfU3gMULKtzNQ2X0qXLhR5tYs0

Image-Adaptive GAN based Reconstruction. arxiv.org/abs/1906.05284

Similarity Problems in High Dimensions. arxiv.org/abs/1906.04842

Edge-Direct Visual Odometry. arxiv.org/abs/1906.04838

This paper evaluates methods in the context of computer vision, specifically when identifying distinct objects in 3D scenes and predicting how far away they are. The new method is called 3D- BoNet.

Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds

paper: https://www.profillic.com/paper/arxiv:1906.01140

Adobe Research and UC Berkeley: Detecting Facial Manipulations in Adobe Photoshop
https://theblog.adobe.com/adobe-research-and-uc-berkeley-detecting-facial-manipulations-in-adobe-photoshop/

ICML Highlight: Contrastive Divergence for Combining Variational Inference and MCMC

https://www.inference.vc/icml-highlight-contrastive-divergence-for-variational-inference-and-mcmc/?fbclid=IwAR1iZXz9zvewdImeZ9lw8BS2a9gk4U7enTYf6x9_pYnZpAhOBO7GILWGiBM

ICML 2019 top papers and highlights https://amicki.co/2019/06/14/ai-weekly-icml-2019-top-papers-and-highlights/

Get SMPL-X, an expressive 3D body that extends the popular SMPL body model with an expressive face and articulated hands. Use SMPLify-X to estimate SMPL-X from a single image. This appears at CVPR.
Project: https://smpl-x.is.tue.mpg.de/
Video: https://www.youtube.com/watch?v=XyXIEmapWkw&feature=youtu.be
Code: https://lnkd.in/dvPDjkF

Semantic Image Synthesis with Spatially-Adaptive Normalization
paper : https://arxiv.org/abs/1903.07291

* code : https://github.com/taki0112/SPADE-Tensorflow

Integrate logic and deep learning with #SATNet, a differentiable SAT solver! #icml2019

Paper: https://arxiv.org/abs/1905.12149
Code: https://github.com/locuslab/SATNet

NIPS 2017 Invited talk "Deep Reinforcement Learning with Subgoals"
By David Silver: https://vimeo.com/249557775
#ArtificialIntelligence #DeepLearning #MachineLearning #NeuralNetworks #ReinforcementLearning

Deep Learning: AlphaGo Zero Explained In One Picture
By L.V.: https://api.ning.com/files/G3detyndwpXvT8Py3CFA1rtuPS549-KcvNCPjfyaORlWtrBVjnT7MSsnV5zQmlOYZg8n9cIqQqf2u4GMq0VHnN1AE-nlYFnx/porc.png
#AlphaGo #ArtificialIntelligence #DeepLearning #NeuralNetworks #ReinforcementLearning