Monocular world meshing
AssigneeSnap Inc.
InventorsErick Mendez Mendez, Isac Andreas Müller Sandvik, Qi Pan, Edward James Rosten, Andrew Tristan Spek, Daniel Wagner, Jakob Zillner
Life of this patent
Estimating the term…
Abstract
A depth estimation system to perform operations that include: receiving image data generated by a client device, the image data comprising a depiction of an environment; identifying a set of image features based on the image data; determining a pose of the client device based on the set of features; generating a depth estimation based on the image data and the pose of the client device; and generating a mesh model of the environment based on the depth estimation.
Claims20
A method comprising:
- receiving an image that comprises image data generated by a client device from a single capture, the image data comprising a depiction of an environment;
- identifying a set of image features within the image based on the image data;
- determining a pose of the client device based on the set of features from within the image;
- providing the image data and the pose of the client device to a machine-learned (ML) model;
- receiving, from the ML model, a depth estimation based on the image data and the pose of the client device;
- and
- generating a mesh model of the environment based on the depth estimation.
A system comprising:
- a memory;
- and
- at least one hardware processor coupled to the memory and comprising instructions that causes the system to perform operations comprising:
- receiving an image that comprises image data generated by a client device from a single capture, the image data comprising a depiction of an environment;
- identifying a set of image features within the image based on the image data;
- determining a pose of the client device based on the set of features from within the image;
- providing the image data and the pose of the client device to a machine-learned (ML) model;
- receiving, from the ML model, a depth estimation based on the image data and the pose of the client device;
- and
- generating a mesh model of the environment based on the depth estimation.
A non-transitory machine-readable storage medium comprising instructions that, when executed by one or more processors of a machine, cause the machine to perform operations comprising:
- receiving an image that comprises image data generated by a client device from a single capture, the image data comprising a depiction of an environment;
- identifying a set of image features within the image based on the image data;
- determining a pose of the client device based on the set of features from within the image;
- providing the image data and the pose of the client device to a machine-learned (ML) model;
- receiving, from the ML model, a depth estimation based on the image data and the pose of the client device;
- and
- generating a mesh model of the environment based on the depth estimation.
Dependencies are inferred from the claim text.