Revisiting Faster R-CNN: A Deeper Look at Region Proposal Network

Guangxing Han, Xuan Zhang, Chongrong Li

Abstract:

Currently, state-of-the-art object detectors are based on
Faster R-CNN. We firstly revisit Faster R-CNN and explore problems in
it, e.g., coarseness of feature maps for accurate localization, fixed-window
feature extraction in RPN and insensitivity for small scale objects. Then
a novel object detection network is proposed to address these problems.
Specifically, we utilize a two-stage cascade multi-scale proposal genera
tion network to get high accurate proposals: an original RPN is adopted
to initially generate coarse proposals, then another network with multi
layer features and RoI pooling layer are introduced to refine these propos
als. We also generate small scale proposals in the second stage simulta
neously. After that, a detection network with multi-layer features further
classifies and refines proposals. A novel 3-step joint training algorithm is
introduced to optimize our model. Experiments on PASCAL VOC 2007
and 2012 demonstrate the effectiveness of our network.