A mobile device stores a target image and target image data collected contemporaneously with the target image. The mobile device receives a reference position indication that corresponds to the target image data and receives a video feed from a camera while the mobile device is in the reference position. The mobile device detects a match between a first image from the video feed and the target image, unlocks an augmented reality space, and instructs presentation of a virtual object within the augmented reality space. The mobile device receives sensor data and a continuing video feed from the camera, compares a second image from the continuing video feed with the first image, and identify common features in the first and second images. The mobile device detects a location change based on the sensor data and changes in the common features between the first and second images.