5,337
edits
Line 549: | Line 549: | ||
'''Exam Question''' | '''Exam Question''' | ||
Benefits: | |||
* Removes SVM part to do end-to-end training. | |||
* Each region is cropped from a feature map of the image rather than the raw image. | |||
** One con is the feature map is lower-res so small objects may become <math>1\times1</math> features. | |||
** '''RCNN is better for smaller objects''' | |||
===Faster R-CNN=== | |||
Focuses on the region proposals by replacing selective search with a region proposal network. | |||
Computes region proposals on-the-fly. | |||
Contains the following | |||
# Feature extractor | |||
# RoI Proposal Network | |||
# RoI Classification & Regression Network | |||
;How region proposal works | |||
# Given an image, pass through conv filters to get feature maps. | |||
# Map each pixel to <math>k</math> anchor boxes. | |||
# Then a layer outputs foreground-background classification and another outputs bounding box regression for each pixel. | |||
Generally two-stage models perform better. Everything is trained end-to-end. | |||
==Will be on the exam== | ==Will be on the exam== |