Fast R-CNN
Improvements w.r.t. R-CNNare:
ROI Pooling: to extract an equal-length vector from all the proposed regions
Single stage model
Shares the CNN layers across all regions, thanks to the ROI Pooling.
No disk caching
Higher accuracy
The feature map from the last convolutional layer is fed to an ROI Pooling layer to extract a fixed-length vector from each region. ROI Pooling splits the region in a grid and applies max pooling on each cell.
For further info, check the ROI Pooling section.
The ROI Pooling output is fed into a FC layer that splits into two branches:
Softmax layer to predict class scores
FC layer with a regression head to predict bounding boxes
Drawbacks
Depends on Selective Search -> very slow
Last updated
