ObjectCount by Flutter and Tensorflow lite (or whatever ML model you suggested)
Budget: $250 – $750 USD
So, this is a Front-end project. No backend component is required.
Use Case:
A single web page application can get the video stream from the device's camera and use the ML model to label the object the user selects. There is a count number at the bottom of the screen. When the object completely crosses the line in the middle of the screen, it adds 1 to the counter. The counter is reset to zero when the app restart or users press the reset button.
Layout:
Please refer to the attachment. Use the default flutter framework for all components, including input boxes, buttons, etc.
Object to label:
Car, people
Implementation Framework:
Flutter and Tensorflow lite (or whatever ML model you suggested)
Run time Compatibility:
The application should work on Windows and Mac OS computers with all modern chromium browsers, including Chrome, Firefox, Edge, and Safari.
For mobile devices, it should work on Chrome on Android devices and Safari on iOS devices.
Camera Compatibility:
When the application runs on PC/Laptop computers, it should be able to use the video from the USB camera.
On mobile devices, it should use the rear camera.
Develiable:
1. workable software that passes the UAT
2. Source code with readme to document how to build the app in the local development environment.
Use Case:
A single web page application can get the video stream from the device's camera and use the ML model to label the object the user selects. There is a count number at the bottom of the screen. When the object completely crosses the line in the middle of the screen, it adds 1 to the counter. The counter is reset to zero when the app restart or users press the reset button.
Layout:
Please refer to the attachment. Use the default flutter framework for all components, including input boxes, buttons, etc.
Object to label:
Car, people
Implementation Framework:
Flutter and Tensorflow lite (or whatever ML model you suggested)
Run time Compatibility:
The application should work on Windows and Mac OS computers with all modern chromium browsers, including Chrome, Firefox, Edge, and Safari.
For mobile devices, it should work on Chrome on Android devices and Safari on iOS devices.
Camera Compatibility:
When the application runs on PC/Laptop computers, it should be able to use the video from the USB camera.
On mobile devices, it should use the rear camera.
Develiable:
1. workable software that passes the UAT
2. Source code with readme to document how to build the app in the local development environment.