1 minute read

CV & ML - Visual Understanding Beyond Object Recognition

No BiAS Podcast Episode 5

“Vision begins with the eyes but truly takes place in the brain” Fei Fei Li, Director of Stanford AI Lab and Stanford Vision Lab TedTalk

 

 

The main objective of computer vision is to give machines the gift of sight as well as the capacity to understand visual input. This has proven a much more complex task than expected. We often take for granted the fact that our vision evolved biologically over millions of years, and we develop our ability to interpret and classify the world around us in early childhood.

Leveraging BigData for Computer Vision

Initially computer vision was tackled through rules based programing. But we don’t learn to recognize objects through rules, but by examples.

In 2007 Fei Fei Li, the Director of Stanford AI Lab came up with the idea to let the model learn by real world examples, just as humans do, leveraging BigData to train algorithms. She and her team launched ImageNet and compiled a database of 15 million images across 22,000 classes of objects organized by everyday words. Almost 50,000 workers from 167 countries helped clean, sort and label nearly a billion candidate images. This global effort paid off and was a major turning point in computer vision and machine learning.

In the Episode

In this episode Saurabh, Nikhil, and Melody talk through the emergence of computer vision as a discipline, the differences in the way that humans and computers “see” images, and the math behind the algorithms. Then they look at some examples of how computer vision is employed in everyday life including Snapchat filters, YouTube video buffering, medical diagnosis of x-rays, and the use of geospatial mapping for agriculture.

Listen Now