Transformers improve feature matching in images
Labeling TV video with Google reverse image search
DCVNet for improved optical flow pixel tracking in video
Facebook AI's MADGRAD optimizer improves neural network training
Turn Your ConvNet into a Zero-Shot Learner
Fast interactive video segmentation to replace tedious pixel labeling of video
Training Computer Vision Transformers without Natural Images
Google Reinforcement Learning uses successful examples instead of tricky reward functions
Google's MoViNets: Mobile Video Networks for Efficient Video Recognition
Tiny four-bit computers may enable neural nets to be trained directly on smartphones
PyTorch implementation of Yann LeCun Barlow Twins: Self-Supervised Learning via Redundancy Reduction
Use PixelLib to extract objects in images and videos with 5 lines of code
Use PyCaret to apply 14 ML models at once and explore data
Top 10 Computer Vision Papers in 2020
GPT-3 AI can write a passing college paper in 20 minutes
Georgia Tech and Facebook Reduce Size of Deep Learning Recommendation Models by Factor of 112
Google Model Search AutoML automatically optimizes and identifies AI models
Google trained a trillion-parameter AI language model
Machine learning with less than one example per class
New Kaggle survey on ML and Data Science