Object Detection System for Desktop Screenshots using Natural Language Descriptions

We are seeking a skilled freelancer to develop an efficient object detection system for desktop screenshots based on natural language descriptions. The system should accurately identify objects within the screenshots and provide their respective coordinates. It will be helpful to send cursor clicks to specific coordinates given a description of the destination, in an automated way. For the purposes of development, any screenshots will do just fine. Requirements: 1. Language Preference: The system should be developed using either Python 3 or Node.js for compatibility with our existing infrastructure. 2. Natural Language Processing: The system should utilize natural language processing techniques to interpret user-provided descriptions and extract relevant keywords. 3. Image Processing: The system should process desktop screenshots to prepare them for object detection, including resizing and normalization. 4. Object Detection: The system should accurately detect objects within the screenshots based on the provided natural language descriptions. 5. Coordinate Extraction: The system should provide precise coordinates (bounding boxes) for the identified objects within the screenshots. 6. Efficiency and Performance: The system should be optimized for efficiency and speed, considering memory usage and computational resources. Deliverables: - Source code of the object detection system implemented in either Python 3 or Node.js. - Documentation explaining the system's architecture, setup instructions, and usage guidelines. - Sample screenshots and corresponding natural language descriptions for testing and validation. We look forward to receiving your proposals and discussing further details!