Container Localisation and Mass Estimation with an RGB-D Camera

TL;DR

This paper presents a method using an RGB-D camera to locate containers and estimate their empty mass, aiding human-robot interaction by providing key information for grasping, despite occlusions and lighting variations.

Contribution

A novel single-camera approach for container localization and mass estimation that is robust to occlusions and lighting changes, focusing on empty container mass prediction.

Findings

Achieved 71.08% accuracy in mass estimation on the CORSMAL dataset.

Method effectively handles occlusions and lighting variations.

Averages multiple predictions for robust estimation.

Abstract

In the research area of human-robot interactions, the automatic estimation of the mass of a container manipulated by a person leveraging only visual information is a challenging task. The main challenges consist of occlusions, different filling materials and lighting conditions. The mass of an object constitutes key information for the robot to correctly regulate the force required to grasp the container. We propose a single RGB-D camera-based method to locate a manipulated container and estimate its empty mass i.e., independently of the presence of the content. The method first automatically selects a number of candidate containers based on the distance with the fixed frontal view, then averages the mass predictions of a lightweight model to provide the final estimation. Results on the CORSMAL Containers Manipulation dataset show that the proposed method estimates empty container mass obtaining a score of 71.08% under different lighting or filling conditions.