asukasmallriver的博客

转：manutdzou/KITTI_SSDR-CNN is a state-of-the-art detector that classifies region proposals by a finetuned Caffe model. For the full details of the R-CNN system and model, refer to its

# coding: utf-8# Note: this file is expected to be in {caffe_root}/examples# ### 1. Setupfrom __future__ import print_functionimport numpy as npimport matplotlib.pyplot as pltimport pylabplt.rcPa

1.准备工作1). 下载预训练模型http://cs.unc.edu/~wliu/projects/ParseNet/VGG_ILSVRC_16_layers_fc_reduced.caffemodel将此文件放置在/home/software/caffe/models/VGGNet/目录下，如果原来有，最好备份一下。2). 下载VOC2007数据集wget http://host.robots.

1.ssd_detect.py的使用在caffe/examples/ssd下有很多python版本的目标检测文件可用：ssd_detect.pyssd_pascal_webcam.pyssd_pascal_video.py用python方法进行detect比较简单，但根据目前使用情况来看，速度较慢，在Jetson TX1下，执行# Forward pass.detections = net.f