VGG19_py_faster_rcnn
| 
59 | 
0 | 
0
Here is VGG19 end to end training model for Py-Faster-Rcnn. It performs slightly better than VGG16.
Copy models folder to your Py-Faster-Rcnn directory.
Download VGG19 model trained on ImageNet for fine tuning.
go to data/imagenet_models
download VGG19_ILSVRC
save it as VGG19.v2.caffemodel
Run training:
cd $FRCN_ROOT./experiments/scripts/faster_rcnn_end2end.sh [GPU_ID] VGG19 pascal_voc
Model that was trained on VOC 2007 trainval is available for download here.
| model | training data | test data | mAP |
|---|---|---|---|
| Faster RCNN, VGG-16 | VOC 2007 trainval | VOC 2007 test | 69.5% |
| Faster RCNN, VGG-19 | VOC 2007 trainval | VOC 2007 test | 70.4% |
| category | mAP VGG16 | mAP VGG19 |
|---|---|---|
| aeroplane | 69.1% | 70.2% |
| bicycle | 78.3% | 79.9% |
| bird | 68.9% | 67.4% |
| boat | 55.7% | 60.1% |
| bottle | 49.7% | 53.5% |
| bus | 77.6% | 76.1% |
| car | 79.7% | 79.9% |
| cat | 85.0% | 86.2% |
| chair | 51.0% | 51.2% |
| cow | 76.1% | 73.6% |
| diningtable | 64.2% | 68.1% |
| dog | 82.0% | 83.6% |
| horse | 80.5% | 80.3% |
| motorbike | 76.2% | 77.5% |
| person | 75.8% | 77.8% |
| pottedplant | 38.4% | 42.9% |
| sheep | 71.3% | 66.9% |
| sofa | 65.4% | 66.9% |
| train | 77.8% | 77.1% |
| tvmonitor | 66.1% | 67.9% |
上一篇：Drone_FasterRCNN
还没有评论，说两句吧！
热门资源
Keras-ResNeXt
Keras ResNeXt Implementation of ResNeXt models...
seetafaceJNI
项目介绍 基于中科院seetaface2进行封装的JAVA...
spark-corenlp
This package wraps Stanford CoreNLP annotators ...
capsnet-with-caps...
CapsNet with capsule-wise convolution Project ...
inferno-boilerplate
This is a very basic boilerplate example for pe...
智能在线
400-630-6780
聆听.建议反馈
E-mail: support@tusaishared.com
资源分类
