Top-features network in gluon-cv Faster R-CNN

blakec · June 25, 2020, 8:44pm

Hi,

I’m looking into the implementation of Faster R-CNN in Gluon-CV and noticed that the feature extraction is split in 2 parts, before and after feature pooling. The second feature extraction step (with top_feature net) of the pooled features is to my knowledge not described in the Faster R-CNN paper written by Shaoqing Ren, Kaiming He, Ross Girshick, and Jian Sun.

Can someone describe why this additional top_feature extractor is added, or give a paper reference describing this step? Is it to simplify training or does it contribute to better detections during inference?

Best,
Blake

Below the link to the above mentioned implementation:

github.com

dmlc/gluon-cv/blob/master/gluoncv/model_zoo/rcnn/faster_rcnn/faster_rcnn.py

"""Faster RCNN Model."""
from __future__ import absolute_import

import os

import mxnet as mx
from mxnet import autograd
from mxnet.gluon import nn
from mxnet.gluon.contrib.nn import SyncBatchNorm

from .rcnn_target import RCNNTargetSampler, RCNNTargetGenerator
from ..rcnn import custom_rcnn_fpn
from ....model_zoo.rcnn import RCNN
from ....model_zoo.rcnn.rpn import RPN


__all__ = ['FasterRCNN', 'get_faster_rcnn', 'custom_faster_rcnn_fpn']


class FasterRCNN(RCNN):

This file has been truncated. show original

Topic		Replies	Views
Faster_rcnn with resnet50 backbone Gluon	0	310	November 16, 2022
GluonCV, Faster RCNN, normalization layers Discussion	7	854	July 4, 2018
Object detection, finetune F-RCNN models Discussion	3	1123	October 14, 2020
Problem with Faster R-CNN training	3	970	July 1, 2019
Predict with pre-trained Faster RCNN models	2	649	October 23, 2019

Top-features network in gluon-cv Faster R-CNN

Related Topics