Image recognition in Delphi
Budget: $30 – $250 USD
I need you to implement in Delphi 12 (or compatible), an image recognition app that is able to perform high level image detection based on its training data. The aim for the system is to guesstimate whether a given image contains a cat, a car or a house for example.
The system consists of two apps:
1) The training system, where I can give the system training data. The training data will be image files, with somehow the desired outputs defined. For example, whether the training image contains a cat, a car or a house. This training system is then run to produce the weights for the neural network based image recognition code.
2) The image recognition code that uses the training data (i.e. the weights) of the training system, that takes an image as an input and outputs its guess what the image contains.
The training part (i.e. part 1) can be implemented in Python or other language (if you want to use other language, you need to discuss that with me before starting), but the image recognition part (i.e. part 2) must be pure Delphi code with no third party DLL files or anything of that kind used. If you want to use any third party libraries, you need to discuss that with me before starting.
The system must be such that it can provide multiple outputs. For example, in case an image contains a cat, and a house, it needs to provide these two as the outputs.
You must provide the training data for the system.
The system consists of two apps:
1) The training system, where I can give the system training data. The training data will be image files, with somehow the desired outputs defined. For example, whether the training image contains a cat, a car or a house. This training system is then run to produce the weights for the neural network based image recognition code.
2) The image recognition code that uses the training data (i.e. the weights) of the training system, that takes an image as an input and outputs its guess what the image contains.
The training part (i.e. part 1) can be implemented in Python or other language (if you want to use other language, you need to discuss that with me before starting), but the image recognition part (i.e. part 2) must be pure Delphi code with no third party DLL files or anything of that kind used. If you want to use any third party libraries, you need to discuss that with me before starting.
The system must be such that it can provide multiple outputs. For example, in case an image contains a cat, and a house, it needs to provide these two as the outputs.
You must provide the training data for the system.