This directory holds (*after you download them*):
- Caffe models pre-trained on ImageNet
- Faster R-CNN models
- Symlinks to datasets
To download Caffe models (ZF, VGG16) pre-trained on ImageNet, run:
```
./data/scripts/fetch_imagenet_models.sh
```
This script will populate `data/imagenet_models`.
To download Faster R-CNN models trained on VOC 2007, run:
```
./data/scripts/fetch_faster_rcnn_models.sh
```
This script will populate `data/faster_rcnn_models`.
In order to train and test with PASCAL VOC, you will need to establish symlinks.
From the `data` directory (`cd data`):
```
# For VOC 2007
ln -s /your/path/to/VOC2007/VOCdevkit VOCdevkit2007
# For VOC 2012
ln -s /your/path/to/VOC2012/VOCdevkit VOCdevkit2012
```
Install the MS COCO dataset at /path/to/coco
```
ln -s /path/to/coco coco
```
For COCO with Fast R-CNN, place object proposals under `coco_proposals` (inside
the `data` directory). You can obtain proposals on COCO from Jan Hosang at
https://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-computing/research/object-recognition-and-scene-understanding/how-good-are-detection-proposals-really/.
For COCO, using MCG is recommended over selective search. MCG boxes can be downloaded
from http://www.eecs.berkeley.edu/Research/Projects/CS/vision/grouping/mcg/.
Use the tool `lib/datasets/tools/mcg_munge.py` to convert the downloaded MCG data
into the same file layout as those from Jan Hosang.
Since you'll likely be experimenting with multiple installs of Fast/er R-CNN in
parallel, you'll probably want to keep all of this data in a shared place and
use symlinks. On my system I create the following symlinks inside `data`:
Annotations for the 5k image 'minival' subset of COCO val2014 that I like to use
can be found at http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/instances_minival2014.json.zip.
Annotations for COCO val2014 (set) minus minival (~35k images) can be found at
http://www.cs.berkeley.edu/~rbg/faster-rcnn-data/instances_valminusminival2014.json.zip.
```
# data/cache holds various outputs created by the datasets package
ln -s /data/fast_rcnn_shared/cache
# move the imagenet_models to shared location and symlink to them
ln -s /data/fast_rcnn_shared/imagenet_models
# move the selective search data to a shared location and symlink to them
# (only applicable to Fast R-CNN training)
ln -s /data/fast_rcnn_shared/selective_search_data
ln -s /data/VOC2007/VOCdevkit VOCdevkit2007
ln -s /data/VOC2012/VOCdevkit VOCdevkit2012
```
没有合适的资源?快使用搜索试试~ 我知道了~
温馨提示
【资源说明】 1、该资源内项目代码都是经过测试运行成功,功能正常的情况下才上传的,请放心下载使用。 2、适用人群:主要针对计算机相关专业(如计科、信息安全、数据科学与大数据技术、人工智能、通信、物联网、数学、电子信息等)的同学或企业员工下载使用,具有较高的学习借鉴价值。 3、不仅适合小白学习实战练习,也可作为大作业、课程设计、毕设项目、初期项目立项演示等,欢迎下载,互相学习,共同进步!
资源推荐
资源详情
资源评论
收起资源包目录
基于Faster-RCNN KITTI数据集的车辆行人检测算法完整源码+说明.zip (168个子文件)
maskApi.c 8KB
nms_kernel.cu 5KB
.gitignore 84B
.gitignore 70B
.gitignore 15B
.gitignore 9B
.gitignore 7B
.gitmodules 131B
maskApi.h 2KB
gpu_nms.hpp 146B
004545.jpg 120KB
000542.jpg 113KB
000456.jpg 103KB
001150.jpg 87KB
001763.jpg 72KB
voc_eval.m 1KB
xVOCap.m 258B
get_voc_opts.m 231B
Makefile 56B
README.md 2KB
README.md 2KB
README.md 780B
README.md 185B
README.md 90B
README.md 66B
train.prototxt 36KB
train.prototxt 10KB
train.prototxt 10KB
test.prototxt 9KB
test.prototxt 9KB
train.prototxt 8KB
train.prototxt 7KB
train.prototxt 7KB
test.prototxt 7KB
test.prototxt 7KB
test.prototxt 7KB
train.prototxt 7KB
test.prototxt 7KB
train.prototxt 6KB
test.prototxt 6KB
train.prototxt 4KB
train.prototxt 4KB
test.prototxt 4KB
train.prototxt 4KB
test.prototxt 4KB
test.prototxt 3KB
solver.prototxt 540B
solver.prototxt 407B
solver.prototxt 403B
solver.prototxt 401B
solver.prototxt 400B
solver.prototxt 398B
solver.prototxt 397B
solver.prototxt 395B
solver.prototxt 395B
solver.prototxt 392B
solver.prototxt 387B
train.prototxt~ 36KB
train.prototxt~ 10KB
test.prototxt~ 9KB
train.prototxt~ 8KB
train.prototxt~ 7KB
test.prototxt~ 7KB
test.prototxt~ 6KB
test.prototxt~ 6KB
test.prototxt~ 3KB
solver.prototxt~ 401B
stage2_fast_rcnn_train.pt 8KB
stage1_fast_rcnn_train.pt 8KB
stage2_rpn_train.pt 8KB
stage1_rpn_train.pt 8KB
faster_rcnn_test.pt 6KB
stage2_fast_rcnn_train.pt 6KB
stage1_fast_rcnn_train.pt 6KB
stage2_fast_rcnn_train.pt 6KB
stage1_fast_rcnn_train.pt 6KB
stage2_rpn_train.pt 5KB
stage1_rpn_train.pt 5KB
rpn_test.pt 5KB
stage2_rpn_train.pt 5KB
stage1_rpn_train.pt 5KB
faster_rcnn_test.pt 5KB
faster_rcnn_test.pt 4KB
rpn_test.pt 3KB
rpn_test.pt 3KB
stage2_fast_rcnn_solver30k40k.pt 408B
stage1_fast_rcnn_solver30k40k.pt 408B
stage1_rpn_solver60k80k.pt 396B
stage2_rpn_solver60k80k.pt 396B
stage2_fast_rcnn_solver30k40k.pt 390B
stage1_fast_rcnn_solver30k40k.pt 390B
stage2_fast_rcnn_solver30k40k.pt 384B
stage1_fast_rcnn_solver30k40k.pt 384B
stage2_rpn_solver60k80k.pt 378B
stage1_rpn_solver60k80k.pt 378B
stage2_rpn_solver60k80k.pt 372B
stage1_rpn_solver60k80k.pt 372B
faster_rcnn_test.pt~ 5KB
cocoeval.py 19KB
coco.py 16KB
共 168 条
- 1
- 2
资源评论
龙年行大运
- 粉丝: 997
- 资源: 3856
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功