The PyTorch Developer will be responsible for enhancing current object detection frameworks by integrating advanced modules such as Spatial Transformer Networks and RFBNet to improve model accuracy and robustness for identifying restricted items within X-ray images used in airport security. The role requires training the model on domain-specific datasets and conducting thorough evaluations using metrics like mean Average Precision, Precision, and Recall. Deliverables include optimized and documented code, step-by-step setup guides, and direct knowledge transfer to support academic defenses. Candidates must demonstrate strong debugging and testing capabilities along with rigorous adherence to research ethics, ensuring original and reproducible work throughout the project timeline.
Responsibilities
- Modify and extend existing Faster R-CNN architecture by integrating Spatial Transformer Networks (STN) and RFBNet modules to enhance feature extraction and detection performance.
- Train the improved neural network model using the provided airport baggage X-ray dataset with proper preprocessing and data augmentation techniques.
- Evaluate the trained model comprehensively by calculating metrics such as mean Average Precision (mAP), Precision, and Recall to assess detection accuracy and reliability.
- Develop clean, well-organized, and extensively commented code to facilitate understanding and future modifications.
- Prepare a detailed setup guide to enable seamless deployment and replication of the model in various environments.
- Conduct a knowledge sharing session to explain the functionality and implementation details of the code clearly to support viva voce examination preparation.
- Ensure all work is original, plagiarism-free, and complies with ethical standards in research and software development.
- Debug and test different components of the model to identify and fix potential issues that could impair performance.
- Apply data preprocessing and tensor transformations efficiently to optimize the training process and network generalization.
- Collaborate in an independent capacity to manage time effectively and meet project deadlines within the three-week timeframe.
Skills
Project: Neural Network Based Detection of Prohibited Items in Airport Baggage X-ray Images I have a research-based FYP. The gap and implementation plan is ready. Work Required: 1. Modify existing Faster R-CNN code to integrate STN and RFBNet modules 2. Train model on X-ray baggage dataset 3. Evaluate using mAP, Precision, Recall 4. Provide clean, commented code and setup guide 5. 1-hour session to explain code for viva prep Skills Needed: PyTorch, Computer Vision, Object Detection, Python Timeline: 2 weeks Note: Only apply if you can explain the code. Plagiarism-free work required.
What is your Competitive Advantage?
Similar Job Titles
Zubair Anjum
Business Excellence Specialist
Nestle Pakistan
Mona Sharif
Senior Team Lead
Konext PVT. LTD
Sehrish Azhar
Motivational Speaker & Blogger
Living Upbeat
Babar Ahmed Khan
Territory Sales Manager
Engro Foods
Aqib Irshad
Head of HR, Training & OD
FINCA
Ali Aziz
CEO
Shopistan
Ayesha Shaukat
Social Media Marketing Manager
Ivoke
Zohaib Ali
Marketing Manager
Digital Online Devices
Usman Habib
Recruitment & Training Executive
ACS Synergy (Pvt) Ltd.
Muzammil Baig
Digital Marketing Manager
IT Services Group
