Development of a C++ - OpenCV - Yolo Application for Multi-Stream Video Processing and Object Detection
Budget: $250 – $750 USD
Seeking a skilled C++ developer with expertise in computer vision to develop an application capable of handling real-time video processing with the capability to support up to 32 RTSP streams simultaneously. The application should include YOLO-based object detection on selected streams and provide functionalities for video recording and resolution adjustment.
Key Requirements:
1. Multi-Stream Support: The application must connect and manage up to 32 RTSP video streams. Performance must be optimized for handling multiple streams simultaneously, with the ability to change grid size to 1 - 4 - 8 - 16 - 32
2. Selective Object Detection: Integrate YOLO (You Only Look Once) for real-time object detection. The application should allow users to enable YOLO detection on up to 4 streams simultaneously while the remaining streams continue as regular video feeds.
3. Video Recording: Record the video feeds with object detection and regular streams, saving them in MP4 format independently for each stream, and can manage the storing hard drive and can deal with multiple hard drives.
4. Resolution Adjustment: Provide an option in each video window to adjust the resolution ranging from 240p (SD) up to 4K (Ultra HD).
5. User Interface: Develop a user-friendly interface that allows:
- Starting/stopping individual video streams.
- Enabling/disabling YOLO detection on selected streams.
- Adjusting video resolution per stream.
- Viewing detection results in real-time.
- Accessing recorded videos.
Technical Specifications:
- Programming Language: C++
- Libraries/Frameworks: OpenCV for video processing, ONNX Runtime for YOLO implementation, and Qt or similar for the user interface.
- Operating Systems: Windows 10 and later, Ubuntu 20.04 and later.
Note: the delivery should be:
1- Full source code
2- List all libraries used and the version number to prepare the user device upon it (like yolov? - opencv - onnxruntime ...etc)
3- Must be workable offline
4- I will not agree for any additional amounts will be added later, so provide all of your service and the total amount in once
Key Requirements:
1. Multi-Stream Support: The application must connect and manage up to 32 RTSP video streams. Performance must be optimized for handling multiple streams simultaneously, with the ability to change grid size to 1 - 4 - 8 - 16 - 32
2. Selective Object Detection: Integrate YOLO (You Only Look Once) for real-time object detection. The application should allow users to enable YOLO detection on up to 4 streams simultaneously while the remaining streams continue as regular video feeds.
3. Video Recording: Record the video feeds with object detection and regular streams, saving them in MP4 format independently for each stream, and can manage the storing hard drive and can deal with multiple hard drives.
4. Resolution Adjustment: Provide an option in each video window to adjust the resolution ranging from 240p (SD) up to 4K (Ultra HD).
5. User Interface: Develop a user-friendly interface that allows:
- Starting/stopping individual video streams.
- Enabling/disabling YOLO detection on selected streams.
- Adjusting video resolution per stream.
- Viewing detection results in real-time.
- Accessing recorded videos.
Technical Specifications:
- Programming Language: C++
- Libraries/Frameworks: OpenCV for video processing, ONNX Runtime for YOLO implementation, and Qt or similar for the user interface.
- Operating Systems: Windows 10 and later, Ubuntu 20.04 and later.
Note: the delivery should be:
1- Full source code
2- List all libraries used and the version number to prepare the user device upon it (like yolov? - opencv - onnxruntime ...etc)
3- Must be workable offline
4- I will not agree for any additional amounts will be added later, so provide all of your service and the total amount in once