Method and system for tracking a state of a camera

Abstract

The invention relates to a method for determining a state xk of a camera at a time tk, the camera comprising a time-of-flight, TOF, camera configured to provide distance data with phase ambiguity, the method comprising: a) receiving an image of a scene of interest captured by the camera at the time tk, wherein the scene comprises a plurality of landmarks having known positions; b) receiving a state estimate of the camera at the time tk, wherein the state estimate comprises an estimate of a pose of the camera; c) determining positions of two or more features in the image; d) obtaining, for the two or more features, distance data from the TOF camera, the distance data being subject to phase ambiguity; e) for each feature, determining a plurality of candidate distance hypotheses corresponding to distances within a measurable range of the TOF camera that are equivalent to the obtained distance data in view of the phase ambiguity; f) for each feature, evaluating a cost for combinations of (i) a candidate landmark selected from the plurality of landmarks and (ii) a candidate distance hypothesis for the feature, wherein the cost is based on a likelihood of observing the feature given the candidate landmark, the candidate distance hypothesis, and the state estimate; g) jointly determining, across the two or more features, (i) a feature-to-landmark mapping and (ii) a selected distance hypothesis for each feature, based on the evaluated costs; and h) determining the state xk of the camera at the time tk based on the feature-to-landmark mapping and the selected distance hypotheses. The invention also relates to a computer program product and to an assembly.

Dependencies are inferred from the claim text.