Method for navigating through a set of images

A system for navigating through a set of images depicting a scenery, said system comprising:

• a processor configured to:
• receive information relating to a subset of images within said set of images, said information comprising metadata of each image within said subset of images, said metadata including a geographical position of an image capturing sensor when acquiring the corresponding image, a viewing angle of the image capturing sensor when acquiring the corresponding image, and image property information, wherein said subset of images is selected from the set of images based upon each image in the subset of images being acquired from a geographical position within a bounding area defined in relation to a start geographical position of navigation and a target geographical position of navigation;
• receive weight information for determining similarities between images within said subset of images, said weight information controlling weights to be provided for determining a plurality of different measures of similarities between images within said subset of images based on different types of metadata;
• calculate a plurality of combined measures of similarity between pairs of images in the subset of images, each combined measure of similarity representing a similarity between the images in a pair of images within said subset of images, said calculating being performed using said metadata and said weight information;
• generate a graph representation, wherein at least one node in the graph representation is associated with an image included in the subset of images and at least one edge in the graph representation is associated with a transition between two images included in the subset of images, and said generating the graph representation is associated with said calculating of the plurality of combined measures of similarity;
• receive an indication of a target geographical position associated with a distance in a direction relative to a viewing angle of a current image;
• and
• determine a best ranked path from a first image in the subset of images to a second image in the subset of images, wherein the first image has a geographical position within a distance from the start geographical position that is smaller than a set threshold, the second image has a geographical position within a distance from the target geographical position that is smaller than a set threshold, and the path comprises a number of transitions between images from the first image via intermediate images to the second image, and wherein the processor is configured to determine said best ranked path including by being configured to:
• calculate a transition measure of a path from the first image to the second image as a sum of the combined measures of similarity between pairs of images included in the transitions and associated with a starting node representing the first image and a target node representing the second image in the graph representation, and select the best ranked path based on a comparison of the transition measures for paths from the first image to the second image;
• and
• a memory coupled to the processor and configured to provide the processor with instructions.

A computer program product for navigating through a set of images depicting a scenery, embodied in a non-transitory computer readable medium and comprising computer instructions for:

• receiving information relating to a subset of images within said set of images, said information comprising metadata of each image within said subset of images, said metadata including a geographical position of an image capturing sensor when acquiring the corresponding image, a viewing angle of the image capturing sensor when acquiring the corresponding image, and image property information, wherein said subset of images is selected from the set of images based upon each image in the subset of images being acquired from a geographical position within a bounding area defined in relation to a start geographical position of navigation and a target geographical position of navigation;
• receiving weight information for determining similarities between images within said subset of images, said weight information controlling weights to be provided for determining a plurality of different measures of similarities between images within said subset of images based on different types of metadata;
• calculating a plurality of combined measures of similarity between pairs of images in the subset of images, each combined measure of similarity representing a similarity between the images in a pair of images within said subset of images, said calculating being performed using said metadata and said weight information;
• generating a graph representation, wherein at least one node in the graph representation is associated with an image included in the subset of images and at least one edge in the graph representation is associated with a transition between two images included in the subset of images, and said generating the graph representation is associated with said calculating of the plurality of combined measures of similarity;
• receiving an indication of a target geographical position associated with a distance in a direction relative to a viewing angle of a current image;
• and
• determining a best ranked path from a first image in the subset of images to a second image in the subset of images, wherein the first image has a geographical position within a distance from the start geographical position that is smaller than a set threshold, the second image has a geographical position within a distance from the target geographical position that is smaller than a set threshold, and the path comprises a number of transitions between images from the first image via intermediate images to the second image;
• said determining comprising:
• calculating a transition measure of a path from the first image to the second image as a sum of the combined measures of similarity between pairs of images included in the transitions and associated with a starting node representing the first image and a target node representing the second image in the graph representation;
• and
• selecting the best ranked path based on a comparison of the transition measures for paths from the first image to the second image.