Semi-supervised pollen grain detection in microscope images. Integration of Faster R-CNN implementation from torchvision with pre-trained TIMM models.

Supervised pollen grain detection in microscope images. Integration of Faster R-CNN implementation from torchvision with pre-trained TIMM models.

This project implements a complete pipeline for airborne pollen forecasting using several time-series models and the framework sktime. The pipeline evaluates both fully pre-trained foundational models (used directly for inference) and non-foundational models that require additional training/fine-tuning on the provided dataset.

🐝 Bee Project: AI-powered research on pollinators. Includes pollen detection using computer vision to monitor pollination activity, and bee-eater sound detection with audio classification to identify threats. Combines image & audio analysis to support bee conservation.