Abstract. In this work, we present a method to uncover shape from webcams "in the wild." We present a variant of photometric stereo which uses the sun as a distant light source, so that lighting direction can be computed from known GPS and timestamps. We propose an iterative, non-linear optimization process that optimizes the error in reproducing all images from an extended time-lapse with an image formation model that accounts for ambient lighting, shadows, changing light color, dense surface normal maps, radiometric calibration, and exposure. Unlike many approaches to uncalibrated outdoor image analysis, this procedure is automatic, and we report quantitative results by comparing extracted surface normals to Google Earth 3D models. We evaluate this procedure on data from a varied set of scenes and emphasize the advantages of including imagery from many months.

Bennu is of particular interest because it is thought to contain primitive solar system materials, and, as such, it may contain some of the building blocks of life: organic molecules. In addition, Bennu is a potentially hazardous asteroid, with a non-zero probability of impacting Earth (Lauretta et al., 2015). Spectral mapping in the context of the OSIRIS-REx mission entails registering data from the onboard spectrometers to the surface of Bennu. The OSIRIS-REx Visible and InfraRed Spectrometer (OVIRS) (Reuter et al., 2017) and OSIRIS-REx Thermal Emission Spectrometer (OTES) (Christensen et al., 2018) are point spectrometers in which everything within the field of view (FOV, or "spot") is integrated. We focus on OVIRS for this study, but our findings are extendable to OTES because it is also a point spectrometer. To map the OVIRS FOVs to the surface of Bennu, we use a shape model. The shape model is a three-dimensional (3D) closed-volume approximate representation of the asteroid, composed of many small triangular facets (Gaskell et al., 2008). The orientations, sizes, and shapes of the facets as an ensemble define the shape of the asteroid, as shown in Figure 1. The mapping process assigns values obtained from the spot observations to the facets of the shape model (Figure 1).