Enhanced pose determination for display device

A system comprising:

• one or more imaging devices;
• one or more processors;
• and
• one or more computer storage media storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
• matching a portion of a current image of a real-world environment with a patch stored by the system, the patch being associated with a first salient point being tracked by the system, the first salient point being included in a prior image of the real-world environment, wherein the matching is usable to identify the first salient point in the current image, and wherein the first salient point represents a first feature of the real-world environment;
• accessing respective descriptors for the first salient point and a second salient point identified in the current image, wherein the second salient point represents a second feature of the real-world environment, and wherein the system stores a descriptor-based map of the real-world environment indicating real-world locations associated with the first feature and the second feature;
• and
• determining a pose associated with the system, the pose being based on the accessed descriptors and the descriptor-based map.

A method implemented by a head-mounted system, the method comprising:

• matching a portion of a current image of a real-world environment with a patch stored by the system, the patch being associated with a first salient point being tracked by the system, the first salient point being included in a prior image of the real-world environment, wherein the matching is usable to identify the first salient point in the current image, and wherein the first salient point represents a first feature of the real-world environment;
• accessing respective descriptors for the first salient point and a second salient point identified in the current image, wherein the second salient point represents a second feature of the real-world environment, and wherein the system stores a descriptor-based map of the real-world environment indicating real-world locations associated with the first feature and the second feature;
• and
• determining a pose associated with the system, the pose being based on the descriptors and the descriptor-based map.

A head-mounted augmented reality display system comprising:

• one or more outwardly-facing imaging devices configured to obtain images of a real-world environment;
• one or more processors, the processors configured to:
• obtain a current image of the real-world environment;
• perform frame-to-frame tracking on the current image, such that patch-based salient points included in a previous image are matched with locations in the current image, each of the salient points representing a respective feature of the real-world environment;
• perform map-to-frame tracking on the current image, wherein map-to-frame tracking comprises matching descriptors for the patch-based salient points with map-based descriptors stored in a descriptor-based map of the real-world environment, the map-based descriptors being associated with real-world locations of features of the real-world environment;
• and
• determine a pose associated with the display device, the pose indicating at least an orientation of the one or more outwardly-facing imaging devices in the real-world environment.