Real-time traffic sign recognition and classification system using hybrid Vision Transformer (ViT) and ResNet34 models. Achieves ~92% accuracy with YOLOv8-labeled data preprocessing and PyTorch-based inference pipeline.
pytorch cnn-classification resnet-34 deep-learning-models computer-vision-opencv vision-transformers ai-agent image-classification-model yolov8-detection traffic-recognition
-
Updated
May 5, 2024 - Jupyter Notebook