mahyarnajibi / fast-rcnn-torch

Fast R-CNN Torch Implementation
MIT License
134 stars 32 forks source link

Training with ResNet Models #15

Open arefiyan opened 7 years ago

arefiyan commented 7 years ago

If we replace AlexNet with a Resnet50 model for training a fast-RCNN network, how should we initialize the ROI pooling layer? I saw for VGG and AlexNet, it's as follows: local ROIPooling = detection.ROIPooling(6,6):setSpatialScale(1/16) How should we calculate the spatial scale for a ResNet50?