https://github.com/mtli/sap

Code for Towards Streaming Perception (ECCV 2020) :car:
https://github.com/mtli/sap

detection forecasting instance-segmentation streaming-perception tracking

Last synced: about 1 month ago
JSON representation

Code for Towards Streaming Perception (ECCV 2020) :car:

• Host: GitHub
• URL: https://github.com/mtli/sap
• Owner: mtli
• License: mit
• Created: 2020-08-21T16:14:27.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2021-04-14T04:05:29.000Z (almost 4 years ago)
• Last Synced: 2024-01-29T16:33:43.223Z (12 months ago)
• Topics: detection, forecasting, instance-segmentation, streaming-perception, tracking
• Language: Python
• Homepage: http://www.cs.cmu.edu/~mengtial/proj/streaming/
• Size: 10.5 MB
• Stars: 94
• Watchers: 8
• Forks: 19
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# sAP — Code for Towards Streaming Perception



![#fc4903](https://via.placeholder.com/15/fc4903/000000?text=+) ECCV Best Paper Honorable Mention Award

![#fcc203](https://via.placeholder.com/15/fcc203/000000?text=+) Feb 2021: Announcing the [Streaming Perception Challenge](https://eval.ai/web/challenges/challenge-page/800/overview) (CVPR 2021)!

This repo contains code for our ECCV 2020 [**paper**](https://arxiv.org/abs/2005.10420) (Towards Streaming Perception). sAP stands for streaming Average Precision.

The dataset used in this project (Argoverse-HD) can be found on the [**project page**](http://www.cs.cmu.edu/~mengtial/proj/streaming/).

Apr 2021: Note that the code has been updated to match the setting in the streaming perception challenge (mostly about working with newer dependencies). To reproduce the numbers in our paper, please check out the [ECCV paper branch](../../tree/ECCV-paper).

## Contents

- Offline detection

- Streaming (real-time online) detection

- Streaming tracking \& forecasting

- Simulated streaming detection, tracking, \& forecasting

- Simulated streaming detection, tracking, \& forecasting with infinite GPUs 

- Meta-detector Streamer

- Streaming evaluation

- Single-frame schedule simulator

- Helper functions for visualization



## Getting started

1. Follow the instructions [here](doc/data_setup.md) to download and set up the dataset.

1. Follow the instructions [here](doc/code_setup.md) to install the dependencies.

1. Check out the examples to run various tasks in `exp/*`. The documentation for these tasks can be found [here](doc/tasks.md).

## Citation

If you use the code or the data for your research, please cite the paper:

```

@article{Li2020StreamingP,

  title={Towards Streaming Perception},

  author={Li, Mengtian and Wang, Yuxiong and Ramanan, Deva},

  journal={ECCV},

  year={2020}

}

```

## Acknowledgement

We would like to thank the [mmdetection](https://github.com/open-mmlab/mmdetection) team for implementing so many different detectors in a single awesome repo with a unified interface! This greatly reduced our efforts to evaluate different detectors under our streaming setting.