Great article on image enhancing (without NN!!!!)
https://sites.google.com/view/handheld-super-res/
https://sites.google.com/view/handheld-super-res/
Google
Handheld Multi-Frame Super-Resolution
We present a multi-frame super-resolution algorithm that supplants the need for demosaicing in a camera pipeline by merging a burst of raw images. In the above figure we show a comparison to a method that merges frames containing the same-color channels…
Good overview article for 3D pose estimation
https://blog.nanonets.com/human-pose-estimation-3d-guide/
https://blog.nanonets.com/human-pose-estimation-3d-guide/
Nanonets
Intelligent document processing with AI | Nanonets
AI-based intelligent document processing with Nanonets' self-learning OCR. Automate data capture from invoices, receipts, passports, ID cards & more!
Model optimization with new Tensorflow tool
https://medium.com/tensorflow/tensorflow-model-optimization-toolkit-pruning-api-42cac9157a6a
https://medium.com/tensorflow/tensorflow-model-optimization-toolkit-pruning-api-42cac9157a6a
Medium
TensorFlow Model Optimization Toolkit — Pruning API
Since we introduced the Model Optimization Toolkit — a suite of techniques that developers, both novice and advanced, can use to optimize…
Google creating next level translation app, Star Trek translator is not so far as we though)
https://www.technologyreview.com/s/613559/google-ai-language-translation/
https://www.technologyreview.com/s/613559/google-ai-language-translation/
MIT Technology Review
Google’s AI can now translate your speech while keeping your voice
Researchers trained a neural network to map audio “voiceprints” from one language to another.
Unsupervised Learning with Graph Neural Networks
Videos from workshop https://www.ipam.ucla.edu/programs/workshops/workshop-iv-deep-geometric-learning-of-big-data-and-applications/?tab=schedule
Slides: https://helper.ipam.ucla.edu/publications/glws4/glws4_15546.pdf
Videos from workshop https://www.ipam.ucla.edu/programs/workshops/workshop-iv-deep-geometric-learning-of-big-data-and-applications/?tab=schedule
Slides: https://helper.ipam.ucla.edu/publications/glws4/glws4_15546.pdf
IPAM
Workshop IV: Deep Geometric Learning of Big Data and Applications - IPAM
Pytorch implementation of Augmented Neural ODEs
https://arxiv.org/abs/1904.01681
https://github.com/EmilienDupont/augmented-neural-odes
https://arxiv.org/abs/1904.01681
https://github.com/EmilienDupont/augmented-neural-odes
arXiv.org
Augmented Neural ODEs
We show that Neural Ordinary Differential Equations (ODEs) learn representations that preserve the topology of the input space and prove that this implies the existence of functions Neural ODEs...
Interesting new model, faster and smaller the all before
https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
https://github.com/tensorflow/tpu/tree/master/models/official/efficientnet
GitHub
tpu/models/official/efficientnet at master · tensorflow/tpu
Reference models and tools for Cloud TPUs. Contribute to tensorflow/tpu development by creating an account on GitHub.
Very interesting paper from Google Research. Generating video from first and end frames
https://arxiv.org/pdf/1905.10240.pdf
https://arxiv.org/pdf/1905.10240.pdf
From Ian Goodfellow https://twitter.com/goodfellow_ian/status/1133528189651677184
"1/7 Do word embeddings really say that man is to doctor as woman is to nurse? Apparently not. Check out this thread for a description of a short paper I co-wrote with Malvina Nissim and Rob van der Goot, available here: (link: https://arxiv.org/abs/1905.09866) arxiv.org/abs/1905.09866 #NLProc #bias"
"1/7 Do word embeddings really say that man is to doctor as woman is to nurse? Apparently not. Check out this thread for a description of a short paper I co-wrote with Malvina Nissim and Rob van der Goot, available here: (link: https://arxiv.org/abs/1905.09866) arxiv.org/abs/1905.09866 #NLProc #bias"
#vacancy #job #python
Big video analytic company are looking for Data Engineer and Infrastructure Engineer.
Data: 10+Pb + 10Tb/day
Sallary: $4k-6k + shares after 1 year
Office: Kyiv, Gulliver
Description: https://bit.ly/2HOeEaW
Contact: https://www.facebook.com/taras.shumyk
Big video analytic company are looking for Data Engineer and Infrastructure Engineer.
Data: 10+Pb + 10Tb/day
Sallary: $4k-6k + shares after 1 year
Office: Kyiv, Gulliver
Description: https://bit.ly/2HOeEaW
Contact: https://www.facebook.com/taras.shumyk
Google Docs
Data Engineer Position
"Gauge Equivariant Convolutional Networks and the Icosahedral CNN
" pretty interesting way of thinking, i like that
https://arxiv.org/pdf/1902.04615.pdf
" pretty interesting way of thinking, i like that
https://arxiv.org/pdf/1902.04615.pdf
Collection of pretrained models for PyTorch
https://github.com/rwightman/pytorch-image-models
https://github.com/rwightman/pytorch-image-models
GitHub
GitHub - huggingface/pytorch-image-models: The largest collection of PyTorch image encoders / backbones. Including train, eval…
The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (V...
Notes from Karpathy on common mistakes when training NN
https://karpathy.github.io/2019/04/25/recipe/
https://karpathy.github.io/2019/04/25/recipe/
karpathy.github.io
A Recipe for Training Neural Networks
Musings of a Computer Scientist.
Learning Perceptually-Aligned Representations via Adversarial Robustness
https://arxiv.org/abs/1906.00945
Github: https://github.com/MadryLab/robust_representations
https://arxiv.org/abs/1906.00945
Github: https://github.com/MadryLab/robust_representations
New interesting paper to read, on face generation(faster then GANs)
https://arxiv.org/abs/1906.00446
https://arxiv.org/abs/1906.00446
arXiv.org
Generating Diverse High-Fidelity Images with VQ-VAE-2
We explore the use of Vector Quantized Variational AutoEncoder (VQ-VAE) models for large scale image generation. To this end, we scale and enhance the autoregressive priors used in VQ-VAE to...
In case that you didn't read that awesome research about wellness. So if you asked yourself why, that's real answer
https://arxiv.org/abs/1802.07068
https://arxiv.org/abs/1802.07068
arXiv.org
Talent vs Luck: the role of randomness in success and failure
The largely dominant meritocratic paradigm of highly competitive Western cultures is rooted on the belief that success is due mainly, if not exclusively, to personal qualities such as talent,...
Learning Unsupervised Video Object Segmentation through Visual Attention
https://www.researchgate.net/publication/332751903_Learning_Unsupervised_Video_Object_Segmentation_through_Visual_Attention
https://www.researchgate.net/publication/332751903_Learning_Unsupervised_Video_Object_Segmentation_through_Visual_Attention
Google open sourced football game environment for training RL agents. You can use image or game state training there.
https://ai.googleblog.com/2019/06/introducing-google-research-football.html
https://ai.googleblog.com/2019/06/introducing-google-research-football.html
research.google
Introducing Google Research Football: A Novel Reinforcement Learning Environment
Posted by Karol Kurach, Research Lead and Olivier Bachem, Research Scientist, Google Research, Zürich The goal of reinforcement learning (RL) is ...
Good summary article about GANs
https://machinelearningmastery.com/resources-for-getting-started-with-generative-adversarial-networks/
https://machinelearningmastery.com/resources-for-getting-started-with-generative-adversarial-networks/
MachineLearningMastery.com
Best Resources for Getting Started With GANs - MachineLearningMastery.com
Generative Adversarial Networks, or GANs, are a type of deep learning technique for generative modeling. GANs are the techniques behind the startlingly photorealistic generation of human faces, as well as impressive image translation tasks such as photo colorization…
Nice paper on disentanglement representation(where each dimension of embedding represent one feature)
https://proceedings.mlr.press/v97/mathieu19a.html
https://proceedings.mlr.press/v97/mathieu19a.html
PMLR
Disentangling Disentanglement in Variational Autoencoders
We develop a generalisation of disentanglement in variational autoencoders (VAEs)—decomposition of the latent representation—characterising it as the fulfilm...
Tons of presentations from Embedded Vision Summit 2019
https://www.embedded-vision.com/platinum-members/embedded-vision-alliance/embedded-vision-training/downloads/pages/may-2019-summit-slides
https://www.embedded-vision.com/platinum-members/embedded-vision-alliance/embedded-vision-training/downloads/pages/may-2019-summit-slides
Edge AI and Vision Alliance
May 2019 Embedded Vision Summit Replay
Keynotes “Making the Invisible Visible: Within Our Bodies, the World Around Us and Beyond,” a Keynote Presentation from the MIT Media Lab June