A Learning Approach for Real-Time Temporal Scene Flow Estimation from LIDAR Data

Abstract

Many autonomous systems require the ability to perceive and understand motion in a dynamic environment. We present a novel algorithm that estimates this motion from raw LIDAR data in real-time without the need for segmentation or model-based tracking. The sensor data is first used to construct an occupancy grid. The foreground is then extracted via a learned background filter. Using the filtered occupancy grid, raw scene flow between successive scans is computed. Finally, we incorporate these measurements in a filtering framework to estimate temporal scene flow. We evaluate our method on the KITTI dataset.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2017
Accession Number
AD1172571

Entities

People

  • Arash K. Ushani
  • Jeffrey M. Walls
  • Ryan M. Eustice
  • Ryan W. Wolcott

Organizations

  • University of Michigan

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Accuracy
  • Algorithms
  • Automation
  • Autonomous Systems
  • Autonomous Vehicles
  • Computer Stereo Vision
  • Computer Vision
  • Detection
  • Detectors
  • Environment
  • Errors
  • Filters
  • High Resolution
  • Image Registration
  • Measurement
  • Point Clouds
  • Robotics

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Neural Network Machine Learning.
  • Vision Science/Vision Psychology/Cognitive Neuroscience.

Technology Areas

  • Autonomy