Featureless 2D-3D Pose Estimation by Minimising an Illumination-Invariant Loss

TL;DR

This paper introduces a method for estimating the full 3D pose of known objects from a single 2D image without training, relying on an illumination-invariant measure to handle varying lighting conditions.

Contribution

It proposes a novel illumination-invariant distance measure for 3D pose estimation that does not require prior training, camera parameters, or feature matching.

Findings

Effective 3D pose estimation under varying lighting conditions

No need for prior training or explicit feature matching

Works with a single static image for full pose recovery

Abstract

The problem of identifying the 3D pose of a known object from a given 2D image has important applications in Computer Vision ranging from robotic vision to image analysis. Our proposed method of registering a 3D model of a known object on a given 2D photo of the object has numerous advantages over existing methods: It does neither require prior training nor learning, nor knowledge of the camera parameters, nor explicit point correspondences or matching features between image and model. Unlike techniques that estimate a partial 3D pose (as in an overhead view of traffic or machine parts on a conveyor belt), our method estimates the complete 3D pose of the object, and works on a single static image from a given view, and under varying and unknown lighting conditions. For this purpose we derive a novel illumination-invariant distance measure between 2D photo and projected 3D model, which is then minimised to find the best pose parameters. Results for vehicle pose detection are presented.