Skip to main content

Why ResNet50 is important

This page is meant to explain our findings from making the jump from v3 to v4 and why using just the LSTM layer was not enough to capture the features in the data.

Why ResNet50?​

The ResNet50 model is a pre-trained model that has been trained on the ImageNet dataset. It is a deep convolutional neural network that has been trained to recognize objects and features in images.

Image Features​

  • Edges and Lines
  • Corners and Key Points
  • Texture Patterns
  • Color Historgrams
  • Gradent Orientation
  • Scale-Invariant Features
  • Deep Learning Features