ROSA: Reconstructing Object Shape and Appearance Textures by Adaptive Detail Transfer

TL;DR

ROSA is an inverse rendering technique that adaptively refines mesh geometry and reconstructs high-resolution surface details from limited images, overcoming resolution and normal map limitations of previous methods.

Contribution

It introduces a novel adaptive mesh refinement and a tile-based high-resolution texture reconstruction method based on a single pre-trained decoder.

Findings

Adaptive mesh resolution improves detail accuracy.

High-resolution textures capture fine surface details.

Method outperforms previous approaches in detail reconstruction.

Abstract

Reconstructing an object's shape and appearance in terms of a mesh textured by a spatially-varying bidirectional reflectance distribution function (SVBRDF) from a limited set of images captured under collocated light is an ill-posed problem. Previous state-of-the-art approaches either aim to reconstruct the appearance directly on the geometry or additionally use texture normals as part of the appearance features. However, this requires detailed but inefficiently large meshes, that would have to be simplified in a post-processing step, or suffers from well-known limitations of normal maps such as missing shadows or incorrect silhouettes. Another limiting factor is the fixed and typically low resolution of the texture estimation resulting in loss of important surface details. To overcome these problems, we present ROSA, an inverse rendering method that directly optimizes mesh geometry with spatially adaptive mesh resolution solely based on the image data. In particular, we refine the mesh and locally condition the surface smoothness based on the estimated normal texture and mesh curvature. In addition, we enable the reconstruction of fine appearance details in high-resolution textures through a pioneering tile-based method that operates on a single pre-trained decoder network but is not limited by the network output resolution.