As a child, I had great fun watching cars passing by and guessing their brands. Now I'm building a neural network to do the same task. Let's see who performs better!
- Leveraging transfer learning with EfficientNet
- Gathering and organizing a dataset comprising images from multiple car brands, ensuring diverse representation
- Using data augmentation, learning rate scheduling to enhance model generalization
- Actively refining the model architecture and hyperparameters to improve accuracy and robustness
This project uses the Stanford Cars Dataset.
Citation:
3D Object Representations for Fine-Grained Categorization
Jonathan Krause, Michael Stark, Jia Deng, Li Fei-Fei
4th IEEE Workshop on 3D Representation and Recognition, at ICCV 2013 (3dRR-13).
Sydney, Australia. Dec. 8, 2013.