An MRF Optimisation Framework for Full 3D Helmholtz Stereopsis
An MRF Optimisation Framework for Full 3D Helmholtz Stereopsis
复制标题
DOI：
10.5220/0007407307250736
复制
发表时间：
2019-02
期刊：
影响因子：
--
通讯作者：
Gianmarco Addari;Jean-Yves Guillemaut
中科院分区：
文献类型：
--
作者：
Gianmarco Addari;Jean-Yves Guillemaut
真实的世界物体的精确3D建模在诸如数字电影制作和文化遗产保护等许多应用中是必不可少的。然而，目前的建模技术依赖于假设来约束问题，有效地限制了可以重建的场景的类别。一个常见的假设是场景的表面反射率是朗伯的或先验已知的。这些约束在实践中很少成立，并导致不准确的重建。Helmholtz Stereopsis（HS）通过引入反射率不可知建模约束来解决这一限制，但在这一领域的先前工作主要限于2.5D重建，仅提供场景的部分模型。相比之下，本文介绍了第一个马尔可夫随机场（MRF）的优化框架，全三维HS。首先，通过从多个视点执行具有可见性约束的2.5D MRF优化并融合不同的输出来获得初始重建。然后，通过使用定制的迭代条件模式（ICM）算法的体积MRF优化，获得一个细化的3D模型。所提出的方法进行评估与合成和真实的数据。结果表明，所提出的全3D优化显著提高了几何和法向精度，能够实现亚毫米级精度。此外，该方法被证明是强大的闭塞和噪声。
Accurate 3D modelling of real world objects is essential in many applications such as digital film production and cultural heritage preservation. However, current modelling techniques rely on assumptions to constrain the problem, effectively limiting the categories of scenes that can be reconstructed. A common assumption is that the scene’s surface reflectance is Lambertian or known a priori. These constraints rarely hold true in practice and result in inaccurate reconstructions. Helmholtz Stereopsis (HS) addresses this limitation by introducing a reflectance agnostic modelling constraint, but prior work in this area has been predominantly limited to 2.5D reconstruction, providing only a partial model of the scene. In contrast, this paper introduces the first Markov Random Field (MRF) optimisation framework for full 3D HS. First, an initial reconstruction is obtained by performing 2.5D MRF optimisation with visibility constraints from multiple viewpoints and fusing the different outputs. Then, a refined 3D model is obtained through volumetric MRF optimisation using a tailored Iterative Conditional Modes (ICM) algorithm. The proposed approach is evaluated with both synthetic and real data. Results show that the proposed full 3D optimisation significantly increases both geometric and normal accuracy, being able to achieve sub-millimetre precision. Furthermore, the approach is shown to be robust to occlusions and noise.