Time of flight motion compensation revisited

Abstract

In this paper, we study motion artifacts that arise in Time-ofFlight imaging of dynamic scenes caused by the sequential nature of the raw image acquisition process used to compute the final depth image. Many methods for compensation of such errors have been proposed to date, but still lack a proper comparison. We bridge this gap by not only evaluating those methods, but also by providing implementations for all of them as a base-line to the community. By exchanging the calibration model necessary for these methods with a model closer to reality we were able to improve the results on all related methods without any loss of performance.

Tackling 3D ToF Artifacts Through Learning and the FLAT Dataset

“…Motion artifacts occur when objects move and ToF raw measurements are captured sequentially. Gottfried et al [13] identify three ways to attenuate them: reduce the number of sequential measurements; detect and correct the regions affected by motion, both in the raw measurement domain and in the depth domain; or estimate the 2D motion fields between raw 2D measurement maps and apply corrective spatial warping. One way to detect affected pixels in raw measurements is by checking the Plus and Minus rules [14,15] that derive from physical constraints on the light measured in a static scene.…”