Transformers improve feature matching in images
339 NLP Google Colab Jupyter notebooks
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
OpenCV on 4"x2" board detects people, face, cars, human pose, text, 3D depth
Amazon Echo, Google Home can measure your heartbeat
Device trains your dog with computer vision and TensorFlow
Amazon Reinforcement Learning Simplifies Training and Deploying Robots
Microsoft Azure Percept brings AI to the edge with new hardware and software platform