DeepPS2: Revisiting Photometric Stereo using Two Differently Illuminated Images 

 Abstract

Photometric stereo, a problem of recovering 3D surface normals using images of an object captured under different lightings, has been of great interest and importance in computer vision research. Despite the success of existing traditional and deep learning-based methods, it is still challenging due to: (i) the requirement of three or more differently illuminated images, (ii) the inability to model unknown general reflectance, and (iii) the requirement of accurate 3D ground truth surface normals and known lighting information for training. In this work, we attempt to address an under-explored problem of photometric stereo using just two differently illuminated images, referred to as the PS2 problem. It is an intermediate case between a single image-based reconstruction method like Shape from Shading (SfS) and the traditional Photometric Stereo (PS), which requires three or more images. We propose an inverse rendering-based deep learning framework, called DeepPS2, that jointly performs surface normal, albedo, lighting estimation, and image relighting in a completely self-supervised manner with no requirement of ground truth data. We demonstrate how image relighting in conjunction with image reconstruction enhances the lighting estimation in a self-supervised setting.

Acknowledgements

Special thanks to the anonymous reviewers for their constructive feedback on our work. We are grateful to Science and Engineering Research Board (SERB), Department of Science and Technology, Govt. of India for providing support through the IMPRINT 2 grant.

Ashish Tiwari is a PhD student at IIT Gandhinagar and a recipient of the Prime Minister Research Fellowship (PMRF) from MHRD, Govt. of India.

Shanmuganathan Raman is the Jibaben Patel Chair on Artificial Intelligence, Associate Professor, IIT Gandhinagar, Gujarat, India.