Uses deep learning techniques for semantic segmentation in autonomous driving, specifically employing the U-Net architecture to perform pixel-wise classification in challenging weather conditions using TensorFlow and the Mendeley DAWN dataset.

Project Objective's:

1. Develop a Semantic Segmentation Model: Implement a U-Net architecture to perform pixel-wise classification for autonomous driving, enabling accurate scene segmentation under various weather conditions.
2. Enhance Weather Robustness: Analyze and incorporate weather data to improve the model's performance in challenging weather scenarios like fog, rain, and snow, ensuring the model's reliability.
3. Data Preprocessing and Augmentation: Perform comprehensive data preprocessing and apply augmentation techniques to increase the model’s robustness and generalization capability across diverse environments.
4. Evaluate Model Performance: Use evaluation metrics such as Intersection over Union (IoU), accuracy, precision, and recall to assess the effectiveness of the segmentation model in different conditions, focusing on weather-adaptive capabilities.