# Siamese-RPN-tensorflow

**Repository Path**: l-j-l/Siamese-RPN-tensorflow

## Basic Information

- **Project Name**: Siamese-RPN-tensorflow
- **Description**: An re-implementation for Siamese-RPN with Tensorflow
- **Primary Language**: Unknown
- **License**: MIT
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 1
- **Created**: 2021-06-12
- **Last Updated**: 2021-06-12

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

# Siamese-RPN-tensorflow
Code for reproducing the results in the following paper:
- [**High Performance Visual Tracking with Siamese Region Proposal Network**](http://openaccess.thecvf.com/content_cvpr_2018/papers/Li_High_Performance_Visual_CVPR_2018_paper.pdf)  
- [Pytorch version](https://github.com/songdejia/siamese-RPN.git) has been available by my classmates.
- Another version [zkisthebest/Siamese-RPN](https://github.com/zkisthebest/Siamese-RPN.git) have lots of bugs.So I have to re-implementation by myself
## Environment
- python=3.6
- tensorflow=1.10
- cuda=9.0

## Downloading VOT2013 Data
- Enter http://data.votchallenge.net/vot2013/vot2013.zip in your browser
- Unzip the file and move to `./data`
## Downloading YouTube-bb Data
- git clone https://github.com/mbuckler/youtube-bb.git
- python3 download.py ./data 12
## Downloading ILSVRC 2015-VID Data
- wget http://bvisionweb1.cs.unc.edu/ilsvrc2015/ILSVRC2015_VID.tar.gz
## Performance

The red box is for tracing
<div align=center><img width="400" height="400" src="https://github.com/makalo/Siamese-RPN-tensorflow/blob/master/visual/test.gif"/></div>

## Visualization for debug

**bbox in detection** 

- red   -- the groundtruth

- black -- bbox with highest score

- other colors -- bbox with scores from second to tenth.

<div align=center><img width="400" height="400" src="https://github.com/makalo/Siamese-RPN-tensorflow/blob/master/visual/170.jpg"/></div>
<div align=center><img width="400" height="400" src="https://github.com/makalo/Siamese-RPN-tensorflow/blob/master/visual/90.jpg"/></div>

## Training and Evaluation

If your data format is the same as VOT 2013, you can run the code directly. If not, you need to change the utils/image_reader.py or convert the data format to VOT format.

### To train Siamese-RPN:
```
python train.py
```
If you want to see if the training is reasonable in the course of training, you can choose to turn on debug.Just change the __init__() in train.py
```
self.is_debug=True
```
This will result in a debug folder where you can see pictures of the training process, with groundtruth in red and box in top 10 scores in other colors.
### To test Siamese-RPN:

**To test series of images like VOT format**

If you want to test a series of images captured from the video, you need to assign new values `img_path`and `img_label` in config.py, which are the files of your image's path and label, respectively. Then execute the following commands
```
python test.py
```
This command will automatically synthesize videos from image sequences, and also synthesize videos from processed images, which are saved in. / data / vedio

**To test a vedio**

If you are testing a video, you need to put the video in./data/vedio. You can run the following command and select the object you want to track in the first frame according to the program prompt at the beginning.
```
python vedio_test.py test.mp4
```
The 'test.mp4' is the name of your vedio
### Model

I will provide the well-trained model in the next few days