A Kendall Shape Space Approach to 3D Shape Estimation from 2D Landmarks

TL;DR

This paper introduces a novel method using Kendall's shape space to reconstruct 3D shapes from single 2D images, effectively addressing the ill-posed problem with limited training data, demonstrated on shark and human models.

Contribution

The paper presents a new Kendall shape space-based approach for 3D shape estimation from 2D landmarks, improving robustness with small training sets.

Findings

Outperforms state-of-the-art methods on shark and human models

Produces plausible 3D shapes with limited training data

More robust than previous approaches in data-scarce scenarios

Abstract

3D shapes provide substantially more information than 2D images. However, the acquisition of 3D shapes is sometimes very difficult or even impossible in comparison with acquiring 2D images, making it necessary to derive the 3D shape from 2D images. Although this is, in general, a mathematically ill-posed problem, it might be solved by constraining the problem formulation using prior information. Here, we present a new approach based on Kendall's shape space to reconstruct 3D shapes from single monocular 2D images. The work is motivated by an application to study the feeding behavior of the basking shark, an endangered species whose massive size and mobility render 3D shape data nearly impossible to obtain, hampering understanding of their feeding behaviors and ecology. 2D images of these animals in feeding position, however, are readily available. We compare our approach with state-of-the-art shape-based approaches, both on human stick models and on shark head skeletons. Using a small set of training shapes, we show that the Kendall shape space approach is substantially more robust than previous methods and results in plausible shapes. This is essential for the motivating application in which specimens are rare and therefore only few training shapes are available.