search

1085052159 / PysotPlus

27 stars 3 forks source link

readme

PySOT_Plus

My English is poor. And the docs may not be complete. I'm busy during this period. I'll improve the documents when I'm not busy. If you have any problems or find any bugs, please open an issue. If you like this project, please give me a little star.

PySOT_Plus adds some contents to pysot to make it more convenient. Thank you very much for sensetime's open source. This is the un-official implementation. Based on their source codes, I have added the following contents.

  1. Introduce the SiamBAN into this source code
  2. Introduce [MIXED PRECISION TRAINING]() into this source code
  3. optimize toolkit to make it support the predicted txts of pytracking, pysot and official website. The txts with separator "," or "space(\ )" or "tab(\t)" works fine.
  4. Introduce [optuna]() to do hyper-param finetune
  5. Add some shells to make it more convenient. Including train.sh(.py), test.sh(.py), eval.sh(.py), union.sh(.py). union.py: the integration of training, testing and evaluation
  6. Add visdom tools to vis Heatmap

Introduction

Evaluation toolkit can support the following datasets:

:paperclip: OTB2015 :paperclip: VOT16/18/19 :paperclip: VOT18-LT :paperclip: LaSOT :paperclip: UAV123 :paperclip: [GOT10K]() :paperclip: [TrackingNet]()

Besides, you can also generate the results figure like Success Plots and Precision Plots. If you want to show the Plots, please install latex.

Model Zoo and Baselines

Please refer to the PySOT Model Zoo. The model is from official PySOT Model Zoo.

Installation

Please find installation instructions for PyTorch and PySOT in INSTALL.md.

Quick Start: Using PySOT

Add PySOT to your PYTHONPATH

export PYTHONPATH=/path/to/pysotplus:$PYTHONPATH

Download models

Download models in PySOT Model Zoo and put the model.pth in the correct directory in experiments

Webcam demo

python tools/demo.py \
    --config experiments/siamrpn_r50_l234_dwxcorr/config.yaml \
    --snapshot experiments/siamrpn_r50_l234_dwxcorr/model.pth
    # --video demo/bag.avi # (in case you don't have webcam)

Download testing datasets

Download datasets and put them into testing_dataset directory. Jsons of commonly used datasets can be downloaded from Google Drive or BaiduYun. If you want to test tracker on new dataset, please refer to pysot-toolkit to setting testing_dataset.

Test&Eval tracker

cd shell
sh test.sh

The testing results will in the current directory(results/dataset/model_name/)

Training :wrench:

See TRAIN.md for detailed instruction.

Getting Help :hammer:

If you meet problem, try searching our GitHub issues first. We intend the issues page to be a forum in which the community collectively troubleshoots problems. But please do not post duplicate issues. If you have similar issue that has been closed, you can reopen it.

:dart:Solution: Run export PYTHONPATH=path/to/pysot first before you run the code.

:dart:Solution: Build region by python setup.py build_ext —-inplace as decribled in INSTALL.md.

References

Contributors

License

If you think this project help you a lot. It's my honor. Please cite this project if you haved used it. I am a little busy, I have no time to finish the docs. If you have any problems about this project, please open an issue.