Project Information

Objective

To explore the application of data manipulation, transfer learning and built a custom deep neural network for a cutlery image dataset and then perform model evaluation to select the hero model to classify cutlery images into four categories - cup, knife, fork, spoon.

Achievements

  • Enhanced model performance through robust input pipelines, data augmentation, and prefetching, achieving 92% classification accuracy.
  • Tech Stack

    Image processing, Data Augmentation, Xception, ImageNet