The project scope is to determine the similarity between object X in an image by comparing it with an existing set of images in which the sample object is served. For example, I have an existing set of images with boats. When we serve new image which might contain a boat or not, the algorithm shall be able to "tell" the similarity % of the new image compared with the image set we have already provided.
Back-end shall be written in Python and be able to run/perform calculations locally on a Raspberry Pi device.
We will need to have a structure of 2 folders, one with accepted criteria, one with unaccepted criteria.
The acceptance probability is between 75%-100%.
The rejected criteria is for objects with under 61%
The probability between 62%-74% shall be sent for human inspection.
The algorithm shall not require internet connection since the "learning" folder is stored locally.
We are open to your inputs and suggestions.
Feel free to write about your experience and past projects. It will help us decide who shall be interviewed.
Your solution to this project shall be as plain and simple as possible since there is only 1 type of object to be recognized and we will provide photos of that object.
36 freelance font une offre moyenne de £14/heure pour ce travail
Hi I have been doing deep learning development since 2017. I have experienced several state of the art models like inception, vgg, resnet and densenet. Lets discuss
Hello, I have enough experience in writing python program and django. I can help you with writing python program. What about talking your app with me? Ok, Thank you.
Dear Employer, I have extensive experience in Django and Python. Please let me know if you are interested and I would be more than happy to assist you. Regards