Towards real-time object recognition and pose estimation in point clouds

TL;DR

This paper introduces a modular pipeline for real-time 6DoF object recognition and pose estimation in point clouds, achieving high accuracy and processing speeds suitable for practical applications.

Contribution

A novel, modular pipeline combining CNN-based object detection, coarse registration, and dense ICP refinement for real-time pose estimation.

Findings

Achieves up to 83% accuracy on RGB-D Scenes dataset.

Object detection at 90 FPS, pose estimation at 14 FPS.

Modular design enables scheduled execution for real-time multitask processing.

Abstract

Object recognition and 6DoF pose estimation are quite challenging tasks in computer vision applications. Despite efficiency in such tasks, standard methods deliver far from real-time processing rates. This paper presents a novel pipeline to estimate a fine 6DoF pose of objects, applied to realistic scenarios in real-time. We split our proposal into three main parts. Firstly, a Color feature classification leverages the use of pre-trained CNN color features trained on the ImageNet for object detection. A Feature-based registration module conducts a coarse pose estimation, and finally, a Fine-adjustment step performs an ICP-based dense registration. Our proposal achieves, in the best case, an accuracy performance of almost 83\% on the RGB-D Scenes dataset. Regarding processing time, the object detection task is done at a frame processing rate up to 90 FPS, and the pose estimation at almost 14 FPS in a full execution strategy. We discuss that due to the proposal's modularity, we could let the full execution occurs only when necessary and perform a scheduled execution that unlocks real-time processing, even for multitask situations.