Accurate Object Pose Estimation Using Depth Only
Open Access
Abstract
Object recognition and pose estimation is an important task in computer vision. A pose estimation algorithm using only depth information is proposed in this paper. Foreground and background points are distinguished based on their relative positions with boundaries. Model templates are selected using synthetic scenes to make up for the point pair feature algorithm. An accurate and fast pose verification method is introduced to select result poses from thousands of poses. Our algorithm is evaluated against a large number of scenes and proved to be more accurate than algorithms using both color information and depth information.Keywords
This publication has 27 references indexed in Scilit:
- Curve Set Feature-Based Robust and Fast Pose Estimation AlgorithmSensors, 2017
- Going Further with Point Pair Features2016
- Learning Analysis-by-Synthesis for 6D Pose Estimation in RGB-D Images2015
- Point Pair Features Based Object Detection and Pose Estimation Revisited2015
- Discriminatively Trained Templates for 3D Object Detection: A Real Time Scalable Approach2013
- Fast object localization and pose estimation in heavy clutter for robotic bin pickingThe International Journal of Robotics Research, 2012
- Vision-guided Robot System for Picking Objects by Casting ShadowsThe International Journal of Robotics Research, 2009
- Using spin images for efficient object recognition in cluttered 3D scenesIEEE Transactions on Pattern Analysis and Machine Intelligence, 1999
- Iterative point matching for registration of free-form curves and surfacesInternational Journal of Computer Vision, 1994
- Watersheds in digital spaces: an efficient algorithm based on immersion simulationsIEEE Transactions on Pattern Analysis and Machine Intelligence, 1991