fix flask python code and add some features
Budget: $30 – $250 USD
I am looking for a skilled developer to fix some bugs and make minor feature improvements to my Flask Python code. I already have the source code, so you will be working with that. Additionally, I have specific technologies and frameworks that I would like this project to be done in. Ideal skills and experience for this job include:
Improved Shape Recognition:
Implement pre-trained object recognition model (like ResNet, Inception, or Mask_RCNN).
Integrate the chosen model into the existing pipeline for shape detection.
Implement post-processing steps to map the model's output to the corresponding shape descriptions.
Enhanced Color Translation:
Expand the existing color dictionary to include more shades and color names.
Incorporate a translation API (like Google Translate) to provide (accurate) translations for colors not covered by the dictionary.
Separation of Arabic and English Text in OCR:
Implement a language detection system, using a library like langdetect, to identify the language of each piece of extracted text.
Develop a method to separate the Arabic text from the English text based on the identified language.
Incorporation of User Feedback for Fine-Tuning:
Update the user interface to allow users to label each paraphrase as true or false.
Develop a system to collect and store these labels, along with the corresponding logo images and paraphrases.
Periodically use this collected data to train or fine-tune the description generation model.
Improved Shape Recognition:
Implement pre-trained object recognition model (like ResNet, Inception, or Mask_RCNN).
Integrate the chosen model into the existing pipeline for shape detection.
Implement post-processing steps to map the model's output to the corresponding shape descriptions.
Enhanced Color Translation:
Expand the existing color dictionary to include more shades and color names.
Incorporate a translation API (like Google Translate) to provide (accurate) translations for colors not covered by the dictionary.
Separation of Arabic and English Text in OCR:
Implement a language detection system, using a library like langdetect, to identify the language of each piece of extracted text.
Develop a method to separate the Arabic text from the English text based on the identified language.
Incorporation of User Feedback for Fine-Tuning:
Update the user interface to allow users to label each paraphrase as true or false.
Develop a system to collect and store these labels, along with the corresponding logo images and paraphrases.
Periodically use this collected data to train or fine-tune the description generation model.