Marker-Less 3d Object Recognition and 6d Pose Estimation for Homogeneous Textureless Objects: An RGB-D Approach.

Nasim Hajari,Irene Cheng,Gabriel Lugo Bustillo,Harsh Sharma

doi:10.3390/s20185098

Abstract

The task of recognising an object and estimating its 6d pose in a scene has received considerable attention in recent years. The accessibility and low-cost of consumer RGB-D cameras, make object recognition and pose estimation feasible even for small industrial businesses. An example is the industrial assembly line, where a robotic arm should pick a small, textureless and mostly homogeneous object and place it in a designated location. Despite all the recent advancements of object recognition and pose estimation techniques in natural scenes, the problem remains challenging for industrial parts. In this paper, we present a framework to simultaneously recognise the object’s class and estimate its 6d pose from RGB-D data. The proposed model adapts a global approach, where an object and the Region of Interest (ROI) are first recognised from RGB images. The object’s pose is then estimated from the corresponding depth information. We train various classifiers based on extracted Histogram of Oriented Gradient (HOG) features to detect and recognize the objects. We then perform template matching on the point cloud based on surface normal and Fast Point Feature Histograms (FPFH) to estimate the pose of the object. Experimental results show that our system is quite efficient, accurate and robust to illumination and background changes, even for the challenging objects of Tless dataset.

Highlights

Recognising an object and estimating its 6d pose simultaneously has received considerable attention in recent years
We perform template matching on the point cloud based on surface normal and Fast Point Feature Histograms (FPFH)
As it was expected and obvious from these plots, a smaller cell and more histogram bins will result in a denser feature space and higher accuracy

Summary

Introduction

Recognising an object and estimating its 6d pose simultaneously has received considerable attention in recent years. Object recognition is the process of classifying the instances of real-world objects in a scene. The process typically extracts distinct features from the Region of Interest (ROI). Feeds the extracted features to a previously trained model to recognise the object’s category. Pose estimation is the process of finding the pose of a known 3d object in the scene with respect to the camera. Transformation matrix, which consists of three translation and three rotation parameters. Applications such as autonomous driving, augmented reality and robotics vision have a very strong need for accurate and efficient recognition and pose estimation approaches

Methods

Results

Conclusion