Skip to content

sek788432/Waymo-2D-Object-Detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Waymo-2D-Object-Detection


Introduction

This is a repo using tensorflow object detection API to detect Waymo open dataset.
Reference: Tensorflow Docs


Requirements

Details are in the notebook.


Dataset


Label Map

item {
    id: 1
    name: 'vehicle'
}

item {
    id: 2
    name: 'pedestrian'
}

item {
    id: 4
    name: 'cyclist'
}

Preprocess

Filter out lidar informations and leverage bounding boxes of:

FRONT, FRONT_LEFT, FRONT_RIGHT, SIDE_LEFT, SIDE_RIGHT cameras.


Train and Evaluate

Fine tune models pre-trained on COCO dataset.

Train and evaluate at the same time and save the best model checkpoint during training.


Results

CenterNet

Taiwan Street

Waymo Image

SSD ResNet

Taiwan Street

Waymo Image

About

Using Tensorflow Object Detection API to detect Waymo open dataset

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published