![](https://ccweb.imgix.net/https%3A%2F%2Fwww.classcentral.com%2Fimages%2Ficon-black-friday.png?auto=format&ixlib=php-4.1.0&s=fe56b83c82babb2f8fce47a2aed2f85d)
Overview
![](https://ccweb.imgix.net/https%3A%2F%2Fwww.classcentral.com%2Fimages%2Ficon-black-friday.png?auto=format&ixlib=php-4.1.0&s=fe56b83c82babb2f8fce47a2aed2f85d)
Learn to predict depths of moving people by observing frozen people. Understand the applications and limitations of existing technologies like SLAM/MVS and traditional stereo triangulation. Develop skills in single-view depth prediction using multi-view supervision and applying it to internet images. Train models on human data and improve depth prediction accuracy. The course uses RGB images for training and teaches how more input leads to better performance. The intended audience for this course includes computer vision enthusiasts and researchers interested in depth prediction techniques.
Syllabus
Intro
Goal
Where Could We Use This?
Existing Technologies
SLAM/MVS
Traditional Stereo Triangulation
Triangulation Here... Not So Good!
Single View Depth Prediction Using Multi-View Supervision
Internet Images Into Data
After Applying MV...
Depth Prediction on MegaDepth... A Lot Less Noisy!
Statues vs People
Mannequin Challenge
Training Data... Now On People
Get The Training Data
Train The Model (Using RGB- Only)
Prediction on Single RGB Image
More The Input Better The Performance
Getting Pseudo-Depth Map
Final Model Output
Application
Taught by
Launchpad