Motion-based feature correspondence
AssigneeGoogle Inc.
InventorsChristian Frueh, Caroline Rebecca Pantofaru
Life of this patent
Estimating the term…
Abstract
A system and method are disclosed for identifying feature correspondences among a plurality of video clips of a dynamic scene. In one implementation, a computer system identifies a first feature in a first video clip of a dynamic scene that is captured by a first video camera, and a second feature in a second video clip of the dynamic scene that is captured by a second video camera. The computer system determines, based on motion in the first video clip and motion in the second video clip, that the first feature and the second feature do not correspond to a common entity.
Claims31
A method comprising:
- identifying, in a first video clip of a dynamic scene that is captured by a first video camera, a first feature;
- identifying, in a second video clip of the dynamic scene that is captured by a second video camera, a second feature;
- synchronizing in time, by a computer system, the first video clip and the second video clip;
- detecting, by the computer system, a first motion of the first feature across two or more first frames of the first video clip and a second motion of the second feature across two or more second frames of the second video clip that correspond to the two or more first frames of the first video clip based on the synchronizing;
- and
- determining, by the computer system, based on the first motion in the first video clip and the second motion in the second video clip, that the first feature and the second feature do not correspond to a common entity, the first motion in the first video clip and the second motion in the second video clip corresponding to one or more common points in time in the first video clip and the second video clip.
An apparatus comprising:
- a memory to store:
- a first video clip of a dynamic scene that is captured by a first video camera;
- and
- a second video clip of the dynamic scene that is captured by a second video camera;
- and
- a processor to:
- identify a first feature in the first video clip;
- identify a second feature in the second video clip;
- synchronize in time the first video clip and the second video clip;
- detect a first motion of the first feature across two or more first frames of the first video clip and a second motion of the second feature across two or more second frames of the second video clip that correspond to the two or more first frames of the first video clip based on synchronizing of the first video clip and the second video clip;
- and
- determine, based on the first motion in the first video clip and the second motion in the second video clip, that the first feature and the second feature do not correspond to a common entity, the first motion in the first video clip and the second motion in the second video clip corresponding to one or more common points in time in the first video clip and the second video clip.
A non-transitory computer-readable storage medium having instructions stored therein, which when executed, cause a computer system to perform operations comprising:
- identifying, in a first video clip of a dynamic scene that is captured by a first video camera, a first feature;
- identifying, in a second video clip of the dynamic scene that is captured by a second video camera, a second feature;
- synchronizing in time, by the computer system, the first video clip and the second video clip;
- detecting, by the computer system, a first motion of the first feature across two or more first frames of the first video clip and a second motion of the second feature across two or more second frames of the second video clip that correspond to the two or more first frames of the first video clip based on the synchronizing;
- and
- determining by the computer system, based on the first motion in the first video clip and the second motion in the second video clip, that the first feature and the second feature do not correspond to a common entity, the first motion in the first video clip and the second motion in the second video clip corresponding to one or more common points in time in the first video clip and the second video clip.
Dependencies are inferred from the claim text.