People Counting & Age Gender Recognition.
Note: Please bid only if you have similar demo app available to ShowCase (Video/Camera/)
Scope Of Work
The purpose of application to count human (male/female, kids/teen/adult/senior) whenever they entered from a retail shop gate.
I am looking for developer having experience of People Counting and Age/Gender detection. I need a good UI interface for the same to do camera calibration from the frontend so that user can set do the same from frontend rather than code every time. Following features we are looking from application.
Developers who have done this kind project will be given preferences. I will need sample App before assigning/awarding project to developer.
Note: Need documentation with source code to manage in future. Though I can assign future development to same developer to manage enhancements etc.
• Python Or C++
• MySQL Database
• Ubuntu 16.04 and 18.04 that can be installed on devices like Raspberry Pi, ODROID XU4 or Intel NUC box (Celeron Dual Core) 2GB RAM. (Application Should be feasible with ARM & x86 devices).
• Application should be Lightweight so that it can run smoothly.
• Should be able to sync data on Web Server.
1. Should count person when person is visible on camera.
2. Should count person only once within an hour time. It means if any person walked in from the gate and comes after 10 minutes should not be counted again.
3. Same person should not be counted multiple times if he/she stands in front of camera.
4. Any person should not be counted which is in exclusion list. This means employee image can be added into exclusion list and should not be counted.
5. This app is to count retail customers in a retail shop so the minimum accuracy should be 90% in case of total counting and for age/gender it should be minimum 80%
6. Application should take direct feed from camera (live feed) and should count data and then store it in the database. Backend process should send data to server every 5 min (Frontend required to change the time). Any data that synced with server should not be transmitted again.
7. Age/Gender and counting should display on video for real time tracking/check.
8. Good UI interface for the same to do camera calibration from the frontend so that user can change the same from frontend rather than code every time.
9. Rest features can be discussed on Chat.