The Algorithm of Real-time Drowning Detection

I am looking for a skilled developer to create an Artificial Neural Network algorithm for real-time drowning detection. The algorithm should have a desired accuracy of over 80%.

Details:
The main goal is to develop a drowning judgment algorithm based on machine learning or others to detect whether a person is drowning, and to automatically identify drowning events or drowning risks by analyzing input video. The main development functions are as follows:
1. The input video can be processed real-timely
2. Determine the number of people appearing in the video, draw a rectangular frame for selection, and identify their ongoing actions (normal or drowning)
3. If it is in a drowning state, an early warning (box color change, voice alarm) is required

Requirements:
- Proficiency in Artificial Neural Network development
- Strong understanding of real-time data processing
- Familiarity with drowning detection systems

Skills:
- Artificial Neural Network development
- Real-time data processing
- Drowning detection system experience

Experience:
- Previous experience developing algorithms with high accuracy rates
- Knowledge of real-time monitoring systems

Please provide examples of previous work and your proposed timeline for completing this project.