0

What is the input size of faster RCNN RPN? I'm using an object detection API of Tensorflow which is using faster RCNN as region proposal network ( RPN ) and Inception as feature extractor ( according to the config file ). The API is using the online approach in prediction phase and detects every input image singly. however, I'm now trying to feed images to the network in the batch manner by use of Tensorflow dataset API. as you know for make batch out of the data, firstly we need to resize all of the images to a same size. I think the best way of resizing the images is to resize them exactly to the input size of faster RCNN to avoid duplicate resizing. Now my question is what is the input size of the faster RCNN RPN? thanks in advance

Arashsyh
  • 609
  • 1
  • 10
  • 16

1 Answers1

3

It depends on the input resolution which was specified in the pipeline config file, in image_resizer. For example, for Faster R-CNN over InceptionV2 trained on COCO dataset, see this config file. The specified resolution is 600x1024.

On a side note, fully convolutional architectures (such as RFCN, SSD, YOLO) don't restrict to a single resolution, i.e. you can apply them on different input resolution without modifying the architecture. But this doesn't mean that the model will be robust to it if you're training on a single resolution.

netanel-sam
  • 1,862
  • 9
  • 14
  • 2
    Thanks a lot, yeah, after reading the paper I found that you're true and RPN is fully convolutional so there is no specific size for input. But, about 600 and 1024 which are mentioned as min resolution and max resolution in the config file, I think it means the network is on it's the best performance if you feed images with shapes neither greater than 1024 nor smaller than 600. I think it doesn't mean 600 * 1024 input size as you mentioned – Arashsyh Nov 24 '18 at 12:02