Looking for a computer vision developer to help with a funded project. Trying to detect vehicles passing through an ROI and track them with cameras placed on the residential city streets from a 30ft/10m height. Need to track through heavy occlusions, such as overlapping cars while they try to park. The goal is to determine parking availability in real time, for metered and unmetered spaces.
Your algorithm will run on a ARM processor, currently the raspberry pi 3, and the camera is the raspberry pi camera (5MP). There is a fisheye lens on the camera to capture the entire FOV. Hopefully, the algorithm can run in at least 5 fps.
I've tried pursuing this myself, but I can't segment the cars properly inside the ROI while there is traffic.
I am familiar with computer vision and current research, so you will be talking to someone who understands the technical aspects of this work.
I am looking for a professional, who can be creative about solving the problems presented by this project. There are many ways to approach this, and I'm hoping your experience will guide us. I look forward to speaking with you!