Our team is a rookie team based out of Seattle. We are looking into using tensor flow lite for our autonomous. We are currently stuck on what to do after the pixel is detected. Is there a way to find the position of the pixel, relative to a point. For example, a team told us that there is an option to find the where the pixel is relative to the center of the camera. We looked into it and could not find anything backing it up and our software team are stuck right now.
Hopefully you started with one of the sample OpModes provided in the FTC software, for Blocks and Java users. Their filenames all start with “ConceptTensorFlowObjectDetection”.
Those samples give telemetry that provides the info you seek. An image and more info are available here at ftc-docs.