GSoC 2025 Proposal – Enhancing QR/Barcode Detection in OpenCV – Swayam Pandey

[Swayam Pandey]

Dear OpenCV GSoC Team,

I hope you're doing well.

I’m Swayam Pandey, a third-year B.Tech student in Computer Science and Engineering with a specialization in Cloud Computing at SRM Institute of Science and Technology, Chennai.

I’ve submitted my GSoC 2025 proposal titled “Enhancing QR/Barcode Detection in OpenCV for Speed and Robustness.”  
The core idea is to strengthen OpenCV’s current detection module by:

- Implementing a multi-scale detection mechanism for higher accuracy  
- Making the system more robust against poor lighting and perspective distortion  
- Optimizing the pipeline using CUDA/OpenCL where applicable  
- Improving unit tests, visual demos, and documentation for community usage

This builds on my prior experience:
- At IIT Indore, I built a malware detection model using a Hybrid ViT-CNN pipeline  
- At IIT Mandi, I explored self-supervised learning using Vision Transformers (I-JEPA)  
- At DRDO, I implemented metallurgical classification systems  
- At Intel Unnati, I created a real-time GPS toll system using containerized microservices  
- Currently, I’m developing a Medical Language Model using BioBERT and ClinicalBERT on DGX H100, targeting diagnosis from clinical notes using advanced NLP evaluation techniques

I’m actively updating my GitHub with prior work and am very passionate about contributing to OpenCV — both during and after the GSoC period.

I’d truly appreciate any feedback on my proposal or suggestions from the mentor team.
  
Swayam Pandey