This week in AI & Machine Learning: Computer vision is changing how we play baseball, using AI to reduce food waste, Boston Dynamics new warehouse robot, pseudo-labeling for semi-supervised learning, and more.
🤖 Artificial Intelligence News:
- Forget Big Data, Computer Vision is the Next Moneyball ← See how computer vision is changing the way we play and think about sports!
- Artificial Intelligence for Reducing Food Waste ← Germany has pledged to cut food waste in half by 2030. Learn how they’re utilizing AI to help with that.
- Boston Dynamics unveils Stretch: a new robot designed to move boxes in warehouses ← This could be a game changer for Boston Dynamics and maybe their first highly practical robot. Read more about my excitement here.
- Artificial Intelligence Is Learning To Categorize And Talk About Art
- AI in Drug Discovery Starts to Live Up to the Hype
- AI spots cell structures that humans can’t
- How A.I.-powered companies dodged the worst damage from COVID
🛠️ Developer Tools & Education:
- Multi-template matching with OpenCV
- Iterated Local Search From Scratch in Python
- How to Learn Artificial Intelligence
- TensorFlow.js Community “Show & Tell” #5
- MADGRAD Optimization Method “A Momentumized, Adaptive, Dual Averaged Gradient Method for Stochastic Optimization”
🎤 Interesting Podcasts & Interviews:
- Drum Roll, Please: AI Startup Sunhouse Founder Tlacael Esparza Finds His Rhythm | NVIDIA
- TWIMLcon Sponsor Series | TWiML
- AI Ethics and the Meaning of Life — Paul Thagard | artists of data science
- An interview with Jagadish Mahendran, 1st place winner of the OpenCV Spatial AI Competition | PyImageSearch
- In the AI of the Beholder | MIT
📄 Notable Research Papers:
- SimPoE: Simulated Character Control for 3D Human Pose Estimation
- Multiview Pseudo-Labeling for Semi-supervised Learning from Video
- LoFTR: Detector-Free Local Feature Matching with Transformers
- Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesis
- Unconstrained Scene Generation with Locally Conditioned Radiance Fields
- Understanding Robustness of Transformers for Image Classification
- CvT: Introducing Convolutions to Vision Transformers
- StyleCLIP: Text-Driven Manipulation of StyleGAN Imagery