Feeling impressed to write down your first TDS put up? We’re always open to contributions from new authors.
Earlier than we get into this week’s number of stellar articles, we’d wish to take a second to thank all our readers, authors, and members of our broader neighborhood for serving to us attain a significant milestone, as our followers depend on Medium simply reached…
We couldn’t be extra thrilled — and grateful for everybody that has supported us in making TDS the thriving, learning-focused publication it’s. Right here’s to extra progress and exploration sooner or later!
Again to our common enterprise, we’ve chosen three latest articles as our highlights this week, centered on cutting-edge instruments and approaches from the ever-exciting fields of pc imaginative and prescient and object detection. As multimodal fashions develop their footprint and use instances like autonomous driving, healthcare, and agriculture go mainstream, it’s by no means been extra essential for knowledge and ML practitioners to remain up-to-speed with the newest developments. (If you happen to’re extra inquisitive about different subjects for the time being, we’ve acquired you lined! Scroll down for a handful of rigorously picked suggestions on neuroscience, music and AI, environmentally acutely aware ML workflows, and extra.)
- Mastering Object Counting in Videos
Correct object detection in movies comes with a number of recent challenges when in comparison with the identical course of in static photos. Lihi Gur Arie, PhD presents a transparent and concise tutorial that exhibits how one can nonetheless accomplish it, and makes use of the enjoyable instance of counting shifting ants on a tree to make her case. - Spicing Up Ice Hockey with AI: Player Tracking with Computer Vision
For anybody searching for a radical and fascinating mission walkthrough, we strongly advocate Raul Vizcarra Chirinos’ writeup of his latest try to construct a hockey-player tracker from (roughly) scratch. Utilizing PyTorch, pc imaginative and prescient strategies, and a convolutional neural community (CNN), Raul developed a prototype that may observe gamers and gather primary efficiency statistics. - A Crash Course of Planning for Perception Engineers in Autonomous Driving
Whereas we would nonetheless be years away from self-driving automobiles dominating our roads, researchers and trade gamers have made important progress lately. Practitioners who’d wish to develop their data of planning and decision-making within the context of autonomous driving shouldn’t miss Patrick Langechuan Liu’s complete “crash course” on the subject.