wadda

Wind's Autonomous Driving Development Art


Keywords
adas, deeplearning
License
MIT
Install
pip install wadda==0.0.1

Documentation

WADDA

Wind's Autonomous Driving Development Art

This repository provides some handy tools and useful libraries

Tools Description
pcd visualizer view point cloud files
gif generator generate gif from image or point cloud
data collection record data via ros
voc2coco convert standard voc format dataset to coco format
ros visualizer convert the common message format of ros to marker for visualization
Libs Description
pypcd libraries for working with point clouds

Install

sudo apt install ros-xxx-foxglove-msgs # xxx can be noetic, melodic, etc.
pip3 install wadda

The following is a brief introduction to its simple usage. For more advanced usage, please refer to its documentation.

Simple Usage

wadda [function_name] [path]

pcd visualizer

  • path : pcd file or the folder containing the pcd file

  • tricks:If viewing a folder containing pcd files, use the "space" and "z" keys to control viewing the next and previous frames, and the "q" key to exit

wadda pcd . # view pcd file or pcd folder

data collection

  • path:Specify the path where you want the data to be stored

  • pro:For advanced usage, please refer to the doc

wadda dc . # By filtering the default ros message type, and then store the data

gif generator

All folders under this path will be traversed. If a folder contains images or point cloud files, a gif will be generated with the name of the folder and stored in its parent directory

  • path:Want to traverse the root path of the generated gif
wadda gif . # generator gif from specify path

voc2coco

  • path:The path where the standard voc format data set is located

    path
    ├── Annotations
    ├── coco
    ├── ImageSets
    ├── JPEGImages
    └── labels.txt
  • labels.txt :The labels.txt file must be included, and its content is the name of the category, which is used to map from class name to label id when converting to coco

    class_name_0
    class_name_1
    class_name_2
    ...
wadda v2c . # convert voc to coco

ros visualizer

  • pro:For advanced usage, please refer to the doc
wadda ros # start ros visualizer for converting ros msg

pypcd

from wadda import pypcd
# parse ros pointcloud2 data
pc = pypcd.PointCloud.from_msg(data)
x = pc.pc_data['x']
y = pc.pc_data['y']
z = pc.pc_data['z']

# parse pcd format file
pc = pypcd.PointCloud.from_path('foo.pcd')
x = pc.pc_data['x']
y = pc.pc_data['y']
z = pc.pc_data['z']