Original Source Here
Daily Newsletter — 20th December 2020
Python Library for Data Annotation, Free Webinar on Few-Shot Learning and AI-based Image Recovery in today’s Data Science Daily 📰
Hover — Python Library for Rapid & Intuitive Data Annotation
Hover is a machine teaching library that enables intuitive and efficient supervision. In other words, it provides a map where you hover over and labels your data differently. For instance, you can:
- 🌱 annotate an intuitively selected group of data points at a time
- 🎡 throw a model in the loop and exploit active learning
- 🐳 cross-check with Snorkel-based distant supervision
Free Webinar on Explainable, Adaptive, and Cross-Domain Few-Shot Learning
This webinar discusses the recent advances in few-shot learning, a regime where only a handful of training examples (maybe just one) are available for learning novel categories unseen during training.
It will cover the following topics too:
- Method for few-shot classification that is capable of matching and localizing instances of novel categories, despite being trained and used with only category level image labels and without any location supervision, also opening the door for weakly supervised few-shot detection.
- Method for meta-learning a model that automatically modifies its architecture to better adapt to novel few-shot tasks.
- Discuss the limitation of the current few-shot learning methods when handling extreme cases of domain transfer, and offer a new benchmark and some ideas towards cross-domain few-shot learning.
Register here: https://www.meetup.com/2d3d-ai/events/275047151
Leonid Karlinsky leads the CV & DL research team in the Computer Vision and Augmented Reality (CVAR) group @ IBM Research AI. Before joining IBM, he served as a research scientist in Applied Materials, Elbit, and FDNA. He is actively publishing and reviewing at ECCV, ICCV, CVPR and NeurIPS, and is serving as an IMVC steering committee member for the past 3 years.
AI Image Upscaling, Deblurring, Denoising and Restoration
A Deep Learning AI model is trained on 100s of millions of images and learns to intelligently increase image resolution without losing quality.
Here the model is trained on 10s of millions of synthetic blurry images and learns to intelligently sharpen them.
A model is trained on 10s of millions of synthetic noisy images and learns to intelligently reduce noise.
Deep Learning model is trained on 100s of thousands of old and damaged photos and learns to intelligently repair them.
Trending AI/ML Article Identified & Digested via Granola by Ramsey Elbasheer; a Machine-Driven RSS Bot