Training Data for Computer Vision at Figure Eight with Qazaleh Mirsharif - TWiML Talk #144

This article discusses Qazaleh Mirsharif's work on computer vision at Figure Eight, focusing on the creation of training datasets for two projects: classifying retinal images for disease diagnosis and detecting parking signs from Google Street View images. The interview highlights the dataset building process and key lessons learned. The article references a previous TWiML Talk (#122) on a similar topic, providing context. It emphasizes the importance of data in machine learning projects, particularly in computer vision applications like medical image analysis and autonomous driving (parking sign detection).

• •The article highlights the importance of training data in computer vision projects.
• •It showcases real-world applications of computer vision, such as medical image analysis and autonomous driving.
• •The interview provides insights into the process of building datasets and the lessons learned from it.