Computer Vision. Exploring integration of Visual Geometry into DL. PyTorch 🍮🔦 Finishing PhD at the University of York 🎓 open to remote work opportunities 👋
Self-Supervised Relative Pose With Homography Model-Fitting in the Loop
Bruce R. Muller, William A. P. Smith
tl;dr: ground plane for homography-based visual odometry
#WACV2023openaccess.thecvf.com/conten…
My kid found a bug in the online homework submission system used by RSM (Russian School of Mathematics). Her answer is certainly correct, but it’s probably not the answer that the question-designers were looking for.
Nicholas @AnthropicAI explains larger models can attain better performance following precise scaling laws. But the compute needed to train these models can only be attained using many coordinated machines that are communicating data between them. Learn: youtube.com/watch?v=qscouq3l…
It’s never been easier to deploy a state-of-the-art ML model to a phone. In this blog, we’ll provide a quick overview of PyTorch Mobile-powered demo apps running various state-of-the-art PyTorch 1.9 ML models spanning images, video, audio and text.
pytorch.org/blog/mobile-demo…
Class-Balanced Loss Based on Effective Number of Samples (Yin Cui, Menglin Jia, Tsung-Yi Lin, Yang Song, Serge Belongie)...
xpaperchallenge.org/cv/surve…
The first 12 mo of a startup can be compressed to:
1. Is there demand & does anyone want this?
2. Is it useful & will they keep using it?
3. How do I get more people to keep using it?
Interspersed w/ periods of irrational conviction & temp sadness despite nothing having changed