Techniques for determining a current location of a mobile device

An apparatus to identify locations comprising:

• a processor component;
• a simultaneous location and mapping (SLAM) candidate component to identify a first candidate key frame that matches a current captured frame to at least a first degree from an interval-based key frame set, the current captured frame captured by a camera of a mobile device of surroundings within a defined area, and the interval-based key frame set comprising key frames selected based on a recurring interval from multiple earlier captured frames captured by the camera during generation of a map of the defined area;
• a convolution neural network (CNN) candidate component to identify a second candidate key frame that matches the current captured frame to at least a second degree from a difference-based key frame set, each key frame of the difference-based key frame set selected from the multiple earlier captured frames based on a degree of difference from all other key frames already included in the difference-based key frame set;
• and
• a position estimation component to determine a current location of the mobile device on the map from estimates of differences between the current location and locations of the first and second candidate key frames on the map.

A computer-implemented method for controlling communications comprising:

• identifying, by a processor circuit, a first candidate key frame that matches a current captured frame to at least a first degree from an interval-based key frame set, the current captured frame captured by a camera of a mobile device of surroundings within a defined area, and the interval-based key frame set comprising key frames selected based on a recurring interval from multiple earlier captured frames captured by the camera during generation of a map of the defined area;
• identifying a second candidate key frame that matches the current captured frame to at least a second degree from a difference-based key frame set, each key frame of the difference-based key frame set selected from the multiple earlier captured frames based on a degree of difference from all other key frames already included in the difference-based key frame set;
• and
• determining, by the processor circuit, a current location of the mobile device on the map from estimates of differences between the current location and locations of the first and second candidate key frames on the map.

An apparatus to identify locations comprising:

• a processor component;
• a simultaneous location and mapping (SLAM) selection component to select each key frame of a interval-based key frame set from multiple earlier captured frames based on a recurring interval, the key frames of the interval-based key frame set captured by a camera of a mobile device of surroundings within a defined area as the mobile device is moved about the defined area to generate a map of the defined area;
• a convolution neural network (CNN) selection component to select each of the key frames of a difference-based key frame set from the multiple earlier captured frames based on a degree of difference from all other key frames already included in the difference-based key frame set;
• and
• a generation component to correlate each key frame of the interval-based key frame set and each key frame of the difference-based key frame set to a corresponding location along a corresponding pathway of the map.

At least one tangible machine-readable storage medium comprising instructions that when executed by a processor circuit, cause the processor circuit to:

• identify a first candidate key frame that matches a current captured frame to at least a first degree from an interval-based key frame set, the current captured frame captured by a camera of a mobile device of surroundings within a defined area, and the interval-based key frame set comprising key frames selected based on a recurring interval from multiple earlier captured frames captured by the camera during generation of a map of the defined area;
• identify a second candidate key frame that matches the current captured frame to at least a second degree from a difference-based key frame set, each key frame of the difference-based key frame set selected from the multiple earlier captured frames based on a degree of difference from all other key frames already included in the difference-based key frame set;
• and
• determine a current location of the mobile device on the map from estimates of differences between the current location and locations of the first and second candidate key frames on the map.