High Accuracy Human and Vehicle Detection
Budget: $750 – $1,500 USD
I need a proficient expert in Object Detection to create a system that can accurately detect humans and vehicles from a series of user-uploaded images.
Key requirements:
- Capable of detecting both humans and vehicles with high-accuracy
- Have strong experience in computer vision, machine learning, and AI algorithms
- Expertise in working with user-originated image inputs
In addition to these, the ability to create robust and efficient models that can handle a large amount of data would be beneficial. This is an intricate project requiring advanced skills in image processing and machine learning. The better the accuracy, the higher the chance you will land this project.
I more so need to detect vehicles. Humans/pedestrians can be a later input.
I need to be able to keep a count of the cars that drive over a certain spot. I have used Python and YOLO and created a test version, but it needs a lot of improvements.
Ultimately, I would like to be able to detect and count cars going different directions. I would like to have multiple lines that I can draw or move around in order to choose where I want the program to count. I have a lot of videos that need to be counted so I will need to be able to apply this to different angles and videos.
Attached is a picture of what I have right now but, it could be improved significantly. For example, the detection sometimes loses the car and recaptures it, and it counts as two cars, I am only able to detect cars with the horizontal line and I am not able to move it easily, I can only count cars on one line and don't have multiple directions of traffic being counted at the same time. These are just some of the difficulties that I would like to have fixed.
If anyone would be able to help with this kind of project it would be greatly appreciated.
Key requirements:
- Capable of detecting both humans and vehicles with high-accuracy
- Have strong experience in computer vision, machine learning, and AI algorithms
- Expertise in working with user-originated image inputs
In addition to these, the ability to create robust and efficient models that can handle a large amount of data would be beneficial. This is an intricate project requiring advanced skills in image processing and machine learning. The better the accuracy, the higher the chance you will land this project.
I more so need to detect vehicles. Humans/pedestrians can be a later input.
I need to be able to keep a count of the cars that drive over a certain spot. I have used Python and YOLO and created a test version, but it needs a lot of improvements.
Ultimately, I would like to be able to detect and count cars going different directions. I would like to have multiple lines that I can draw or move around in order to choose where I want the program to count. I have a lot of videos that need to be counted so I will need to be able to apply this to different angles and videos.
Attached is a picture of what I have right now but, it could be improved significantly. For example, the detection sometimes loses the car and recaptures it, and it counts as two cars, I am only able to detect cars with the horizontal line and I am not able to move it easily, I can only count cars on one line and don't have multiple directions of traffic being counted at the same time. These are just some of the difficulties that I would like to have fixed.
If anyone would be able to help with this kind of project it would be greatly appreciated.