Skip to content

carranza96/waymo-detection-optimization

Repository files navigation

Waymo 2D Object Detection Optimization

TensorFlow 1.15 Python 3.6 Protobuf Compiler >= 3.0

Enhancing 2D object detection by optimizing anchor generation and addressing class imbalance.

This project uses the TensorFlow Object Detection API and proposes several modifications to the standard Faster R-CNN implementation, improving 2D detection over the Waymo Open Dataset:

  • Per-region anchor optimization using genetic algorithms
  • Spatial ROI features in the second-stage Fast R-CNN header network
  • Reduced focal loss to improve performance over minority and difficult instances
  • Ensemble models using non-maximum suppression

Installation

This project depends on several libraries and contains two submodules that have to be installed:

Full details on the installation steps and system requirements can be found at installation.md

Getting started

Scripts

The scripts folder provides ready-to-use shell scripts for many operations:

Example model configuration

An example Faster R-CNN model configuration is provided in the file pipeline.config, using the proposed improvements: anchor optimization, spatial ROI features, and reduced focal loss

Authors

  • Pedro Lara-Benítez - LinkedIn
  • Manuel Carranza-García - LinkedIn
  • Jorge García-Gutiérrez
  • José C. Riquelme

License

This project is licensed under the MIT License - see the LICENSE.md file for details

About

2D Object Detection with Waymo Open Dataset

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published