January 2023 – May 2023
🔬 Marine Trash Detection Project
- Engineered a Python-based solution using YOLOv7 for marine trash detection
- Achieved 90% recall rate in simulated environments
- Optimized ML model performance:
- 40% decrease in detection time
- 80% detection rate on mAP@[.5, .95]
- Conducted rigorous model evaluation using industry-standard metrics
- Leveraged Pandas for advanced model optimization:
- Fine-tuned hyperparameters
- Implemented data augmentation
- Applied transfer learning techniques
- Thrived in an Agile research environment, ensuring seamless project synchronization
💡 Key Achievements:
- Developed cutting-edge object detection system for marine environments
- Significantly improved model efficiency and accuracy
- Contributed to potential environmental impact through AI application