Image-Based Relighting using Room Lighting Basis

CVMP 2016

SIGGRAPH 2015 poster

Antoine Toisoul                      Abhijeet Ghosh

Imperial College London

        lighting_conditionsjpeg    em_projection_basis     bird_graceor_jpeg egg_graceor_jpeg

(a) Room Lighting Basis                           (b) Lighting basis projection                                            (c) Room basis relighting

Fig. 1: Room relighting pipeline. (a) Room lighting basis. (b) Projection of the room lighting basis onto the Grace Cathedral environment (c) Relighting of a wooden bird and a decorative Egg in the Grace Cathedral environment.

relightingor_jpeg                       egg_point_eucalyptus_grove_offset0 egg_point_uffizi_offset0 egg_point_stpeters_probe_offset0 egg_point_pisa_courtyard1024_offset3

                                           (a) Room basis relighting                                                                            (b) Relighting with dense free form sampling

Fig. 2: Comparison of relightings : (a) Relighing using room basis. (b) Relighting using a dense free form acquisition according to Masselus et al. [1]

Abstract: We present a novel and practical approach for image-based relighting that employs the lights available in a regular room to acquire the reflectance field of an object. The lighting basis includes diverse light sources such as the house lights and the natural illumination coming from the windows. Once the data is captured, we homogenize the reflectance field to take into account the variety of light source colours to minimise the tone difference in the reflectance field. Additionally, we measure the room dark level corresponding to a small amount of global illumination with all lights switched off and blinds drawn. The dark level, due to some light leakage through the blinds, is removed from the individual local lighting basis conditions and employed as an additional global lighting basis. Finally we optimize the projection of a desired lighting environment on to our room lighting basis to get a close approximation of the environment with our sparse lighting basis. We achieve plausible results for diffuse and glossy objects that are qualitatively similar to results produced with dense sampling of the reflectance field including using a light stage and we demonstrate effective relighting results in two different room configurations. We believe our approach can be applied for practical relighting applications with general studio lighting.

Publication: Image-Based Relighting using Room Lighting Basis. Antoine Toisoul and Abhijeet Ghosh. European Conference on Visual Media Production (CVMP), Dec. 2016.



Reference: [1] V. Masselus, P. Dutre, and F. Anrys. The free-form light stage. In ACM SIGGRAPH 2002 Conference Abstracts and Applications, SIGGRAPH ’02, pages 262-262, New York, NY, USA, 2002. ACM.

Delicious Twitter Digg this StumbleUpon Facebook