Abstract:Objective: Shape-From-Shading (SFS) is one unique and long-standing problem aiming to reconstruct shape/depth from only one image. The SFS problem can be regarded as an inverse process of image rendering, i.e., solving the Bi-directional Reflection Distribution Function (BRDF) model with the given image. Among various BRDF models, physically based perspective Cook-Torrance BRDF model is generally used because of its universality. However, the solution of Cook-Torrance BRDF model based SFS problem has not been properly addressed because the model is very complicated and non-convex. Method: In this paper, we present a shape and depth joint optimization method under the neural network framework to solve this kind of SFS problem. Our method has three aspects of contributions. Firstly, we remove the unrealistic surface smoothness constraint imposed on the solution of SFS. Secondly, the design of shape and depth joint optimization provides the method with a higher probability of bypassing local optimum and finding the global optimum. Thirdly, the neural network framework enables our method with the ability of performing point-wise optimization, that is, the optimization does not subject to the approximation on the discrete domain. Result: Through experiments, we prove that our method can solve Cook-Torrance BRDF model based SFS problem with good accuracy. The maximum Mean Absolute Error, Mean Relative Error and Root-mean-square error of depth results are 0.8 mm, 1.0% and 1.23 mm respectively, which outperforms related methods.

Shape from Shading Based on Support Vector Regression Algorithm

A Shape-from-shading Method Based on Surface Reflectance Component Estimation.

Shape Representation and Recognition in High Dimensional Feature Space

Training support vector data descriptors using converging linear particle swarm optimization

A Neural-Learning-Algorithm-Based Shape From Shading System

A Machine learning approach for Shape From Shading

New Algorithm of Shape-from-Shading with Specular Reflectance

Fast Algebra Algorithm Of Shape-From-Shading With Specular Reflectance

A Novel Shape from Shading Algorithm for Non-Lambertian Surfaces

A New Shape from Shading Algorithm for Hybrid Surfaces

Research on an Algorithm of Shape from Shading

An Application Of Shape From Shading

New Shape-From-Shading Method With Specular Reflectance Model

A Neural Network Approach for Indirect Shape from Shading

A New Shape from Shading Approach for Specular Surfaces

Interactive Shape from Shading

A Perspective Shape-From-Shading Method Using Fast Sweeping Numerical Scheme

Research on Height Restoration in Shape-from-Shading Method

Optimization of Support Vector Regression Parameters by Flower Pollination Algorithm

Solving Non-Convex Shape-from-Shading Problem Throughneural-Network Based Shape and Depth Jointly Optimization

Solving Shape-From-Shading Problem Through Shape and Depth Joint Optimization