Partial FC is a distributed deep learning training framework for face recognition. + dataset Glint360K, which contains 18 million images of 360K individuals
https://github.com/deepinsight/insightface/tree/master/recognition/partial_fc

Awesome!! https://www.youtube.com/watch?v=86QU7_SF16Q

Free course on Deep Learning with Yann LeCun & Alfredo Canziani
https://atcold.github.io/pytorch-Deep-Learning/

BlazeFace: Sub-millisecond Neural Face Detection on Mobile GPUs

Article: https://arxiv.org/abs/1907.05047

Tensorflow: https://github.com/tensorflow/tfjs-models/tree/master/blazeface

PyTorch:https://github.com/hollance/BlazeFace-PyTorch

Adobe PS added neural filters to their new release)

https://helpx.adobe.com/photoshop/using/neural-filters.html

Breaking news 🔥On November 20th, 2020 Data Science UA will gather participants from all over the world on the 9th Data Science UA Online Conference!
Speakers from TOP companies Amazon, Facebook AI, Airbus, Nvidia, Google, IBM and others are going to share experiences and discuss as much as possible about how AI transforms the world today and what is going to be tomorrow.
What's waiting for you:
✔️ 24 hours non-stop
✔️ 70 speakers from TOP companies
✔️ 3000+ participants
✔️ 3 tracks: Technical track, Workshops track, and Business track
At the conference, you will learn step-by-step algorithms through practical workshops as well as insights which you will be able to bring to life in your own work.
Here’s a 20% OFF for our community - 20_ML_World
💣 Buy the ticket for the best price! We are looking forward to seeing you in November!
Let’s build the Data Science Community together: https://data-science-ua.com/conference/

It's a sad day for all of us, and can dramatically change the way of ML

Creator of Python joins Microsoft https://techcrunch.com/2020/11/12/python-creator-guido-van-rossum-joins-microsoft/

​​С каждым годом количество данных, генерируемых компаниями, растёт всё больше и больше. Но как управлять всеми этими данными, как их эффективно использовать? На этот вопрос сможет ответить Data Scientist, и за это компании готовы платить от 150 000 рублей и больше — ежемесячно!
⠀
Хочешь стать таким же? Онлайн-университет Skillbox может помочь пройти дорогу с нуля до специалиста, ещё и с доступной рассрочкой.
⠀
Тебе предстоит пройти этот путь вместе с опытными специалистами и научиться использовать язык программирования Python, работать с источниками данных CSV, XML и XLSX, познакомиться с анализом данных. Также освоишь машинное обучение, что даст знания в области нейронных сетей. Если нет опыта в программировании, это не помеха, всему обучат и всё расскажут с нуля.

Начать обучение на курсе — ответственное решение, и лучше познакомиться с курсом еще до покупки. Поэтому Skillbox дарит доступ к первым двум модулям курса. На них ты познакомишься с основами языка Python и его применением в анализе данных, научишься пользоваться Jupyter Notebook — рабочим инструментом в области Data Scienсe. Регистрируйся по ссылке: https://clc.to/pa8vIA

Stylized Neural Painting

https://jiupinjia.github.io/neuralpainter/

📃 Paper: https://arxiv.org/abs/2011.08114

📦 Git: https://github.com/jiupinjia/stylized-neural-painting

▶️ Colab: https://colab.research.google.com/drive/1XwZ4VI12CX2v9561-WD5EJwoSTJPFBbr?usp=sharing

Guys, there is 1 day left until the 9th Data Science UA Online Conference, which will take place on November 20, 2020.

Have time to buy your ticket with a 20% discount 😍
Promo code — 20_ML_World

Registration — https://data-science-ua.com/conference/

You will have access to 50 reports and take part in super workshops from Uber, Amazon, and other companies.
* All reports will be available for participants after the end of the Сonference. Data Science UA will send them within a few days after the event.

If you haven't seen the agenda yet, follow the link and find out your favorite speakers' exact speaking time — https://bit.ly/3kmhtAO 💥

Bringing Old Photo Back to Life (CVPR
2020)

http://raywzy.com/Old_Photo/

📃 Paper: https://arxiv.org/abs/2004.09484

📦 Git: https://github.com/microsoft/Bringing-Old-Photos-Back-to-Life

▶️ Colab: https://colab.research.google.com/drive/1NEm6AsybIiC5TwTU_4DqDkQO0nFRB-uA?usp=sharing

gradSLAM: Automagically differentiable SLAM

📃 Paper: https://arxiv.org/abs/1910.10672v3

📦 Git: https://github.com/gradslam/gradslam

U-Net: Going Deeper with Nested U-Structure for Salient Object Detection

📃 https://arxiv.org/abs/2005.09007v2

📦 https://github.com/NathanUA/U-2-Net

#segmentation #unet

Looking for people who worked with writing cuda modules with Halide for Linux. If you are one of them, just post in the comments.

Many different Speech datasets including Ukrainian and Russian
https://github.com/JRMeyer/open-speech-corpora

Good article on "Thin Plate Splines Warping"

https://khanhha.github.io/posts/Thin-Plate-Splines-Warping/

The expansion of AI involves several hot issues related to the workforce (data supply demands, disappearing professions, and undesirable working conditions) that can be overcome with the right crowdsourcing methodology. Nowadays it is important to focus on modern and effective ways to overcome these issues and keep working on critical aspects of successful data collection and labeling.

Yandex Toloka team will present “Crowd Science Workshop: Remoteness, Fairness, and Mechanisms as Challenges of Data Supply by Humans for Automation” at the world’s number one conference on machine learning — NeurIPS 2020. Speakers from all over the world will discuss the future of crowdsourcing markets as well as some topics that were never brought up before:
- Remoteness. A discussion about effectiveness and efficiency of remote work on crowdsourcing platforms.
- Fairness. How the working environment (e.g., a crowdsourcing platform) may help provide executors flexibility in choosing/switching tasks and working hours.
- Mechanisms. Discussion on bilateral mechanisms that not only provide flexibility to the performers, but also guarantee the quality of the result and the efficiency of the process to the customers.

Click the link for more info on Toloka’s workshop at NeurIPS 2020.