# FCHD-Fully-Convolutional-Head-Detector
Code for FCHD - A fast and accurate head detector
This is the code for FCHD - A Fast and accurate head detector. See [the paper](https://arxiv.org/abs/1809.08766) for details and [video](https://youtu.be/gRPA7Hqk3VQ) for demo.
## Dependencies
- The code is tested on Ubuntu 16.04.
- install PyTorch >=0.4 with GPU (code are GPU-only), refer to [official website](http://pytorch.org)
- install cupy, you can install via `pip install cupy-cuda80` or(cupy-cuda90,cupy-cuda91, etc).
- install visdom for visualization, refer to their [github page](https://github.com/facebookresearch/visdom)
## Installation
1) Install Pytorch
2) Clone this repository
```Shell
git clone https://github.com/aditya-vora/FCHD-Fully-Convolutional-Head-Detector
```
3) Build cython code for speed:
```Bash
cd src/nms/
python build.py build_ext --inplace
```
## Training
1) Download the caffe pre-trained VGG16 from the following [link](https://drive.google.com/open?id=10AwNitG-5gq-YEJcG9iihosiOu7vAnfO). Store this pre-trained model in `data/pretrained_model ` folder.
2) Download the BRAINWASH dataset from the [official website](https://www.mpi-inf.mpg.de/departments/computer-vision-and-multimodal-computing/software-and-datasets/). Unzip it and store the dataset in the `data/ ` folder.
3) Make appropriate settings in `src/config.py ` file regarding the updated paths.
4) Start visdom server for visualization:
```Bash
python -m visdom.server
```
5) Run the following command to train the model: `python train.py `.
## Demo
1) Download the best performing model from the following [link](https://drive.google.com/open?id=1DbE4tAkaFYOEItwuIQhlbZypuIPDrArM).
2) Store the head detection model in `checkpoints/ ` folder.
3) Run the following python command from the root folder.
```Shell
python head_detection_demo.py --img_path <test_image_name> --model_path <model_path>
```
## Results
| Method | AP |
| :--------------------------------------: | :---------: |
| Overfeat - AlexNet [1] | 0.62 |
| ReInspect, Lfix [1] | 0.60 |
| ReInspect, Lfirstk [1] | 0.63 |
| ReInspect, Lhungarian [1] | 0.78 |
| **Ours** | **0.70** |
## Runtime
- Runs at 5fps on NVidia Quadro M1000M GPU with 512 CUDA cores.
## Acknowledgement
This work builds on many of the excellent works:
- [Simple faster rcnn pytorch implementation](https://github.com/chenyuntc/simple-faster-rcnn-pytorch) by [Yun Chen](https://github.com/chenyuntc).
- [py-faster-rcnn](https://github.com/rbgirshick/py-faster-rcnn) by [Ross Girshick](https://github.com/rbgirshick).
## Reference
[1] Stewart, Russell, Mykhaylo Andriluka, and Andrew Y. Ng. "End-to-end people detection in crowded scenes." Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.
没有合适的资源?快使用搜索试试~ 我知道了~
Python-快速精准的人头检测器
共33个文件
py:22个
pyc:3个
so:1个
3星 · 超过75%的资源 需积分: 45 76 下载量 167 浏览量
2019-08-09
18:27:02
上传
评论 4
收藏 306KB ZIP 举报
温馨提示
快速精准的人头检测器
资源推荐
资源详情
资源评论
收起资源包目录
Python-快速精准的人头检测器.zip (33个子文件)
FCHD-Fully-Convolutional-Head-Detector-master
data
dataset.py 8KB
util.py 9KB
__init__.py 0B
read_xml.ipynb 2KB
train.py 6KB
precision_recall.png 59KB
LICENSE 2KB
src
head_detector.py 3KB
nms
build.py 279B
non_maximum_suppression.py 6KB
_nms_gpu_post.so 226KB
__init__.pyc 270B
_nms_gpu_post_py.py 720B
_nms_gpu_post.c 294KB
__init__.py 67B
_nms_gpu_post.pyx 970B
non_maximum_suppression.pyc 7KB
_nms_gpu_post_py.pyc 923B
build
temp.linux-x86_64-2.7
_nms_gpu_post.o 344KB
utils.py 7KB
creator_tool.py 5KB
vis_tool.py 8KB
array_tool.py 1KB
bbox_tools.py 10KB
__init__.py 0B
region_proposal_network.py 4KB
config.py 1KB
head_detector_vgg16.py 2KB
__init__.py 0B
trainer.py 5KB
data_convert_hollywood.py 4KB
README.md 3KB
head_detection_demo.py 3KB
共 33 条
- 1
资源评论
- weixin_568565352021-04-03请问哪里官网可以下载呀
- 码农g2020-07-11不用画积分下载,官网上有
- Antilogy2020-09-04网上可以下载到的源码 作者也不说清楚
weixin_39840387
- 粉丝: 780
- 资源: 3万+
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