Giter VIP home page Giter VIP logo

argoverse_baselinetracker's Introduction

[update 04/2020]

The new CVPR2020 argoverse tracking challenge provides detections!

See our new baseline using the provided, improved detections here: https://github.com/johnwlambert/argoverse_cbgs_kf_tracker

Link to download the new detections: https://s3.amazonaws.com/argoai-argoverse/detections_v1.1b.zip

Argoverse Baseline Tracker

Baseline tracker code release for the paper Argoverse: 3D Tracking and Forecasting With Rich Maps, CVPR 2019. [paper]  [website] 

Introduction

This tracker implementation is meant to be a baseline example to demonstrate the use of the map and tracking data in Argoverse dataset. We don't claim its performance to be the best, and we are looking forward to more map-based tracking methods developed using Argoverse in the future.

Requirements

To run this tracker, please install following requirements:

example:

Command for running this tracker tracker on Argoverse:

python3 run_tracking.py --path_dataset=/path/to/argoverse-tracking/test --log_id=<log_id>  --path_output=/path/to/output/folder --use_maskrcnn --region_type=roi --use_map_lane --motion_model='const_v' --measurement_model=both --fix_bbox_size --dataset_name=Argoverse

You can use --save_bev_imgs to print birds-eye-view image as above example or --show_segmentation to plot 3D visualization of segmentation result if mayavi is installed. Tracker output format is the same as Argoverse tracking label format.

Docker Image

It might be tricky to install all the dependencies, so we provided docker image.

  • Docker version 18.09.7, build 2d0083d

To run docker image, first install nvidia-docker and then run following command to build image using the provided DockerFile:

nvidia-docker build -t baselinetracker docker/

After building the image, run following command to start. Mount dataset folder so the data can be accessed in docker environment:

nvidia-docker run -v /path/to/argoverse-tracking/test:/data  -it baselinetracker:latest

And then clone this repo:

git clone https://github.com/alliecc/argoverse_baselinetracker
cd argoverse_baselinetracker
wget https://www.nayuki.io/res/smallest-enclosing-circle/smallestenclosingcircle.py

Here is an example command to start tracker. The tracking output would be stored in /tracking_output.

python3 run_tracking.py --path_dataset=/data --log_id=0f0d7759-fa6e-3296-b528-6c862d061bdd  --path_output=/tracking_output --use_maskrcnn --region_type=roi --use_map_lane --motion_model='const_v' --measurement_model=both --fix_bbox_size --dataset_name=Argoverse

argoverse_baselinetracker's People

Contributors

alliecc avatar

Watchers

 avatar

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.