Google & Microsoft have both released API's that will give us access to technology that can recognise content within in photos - so we are exploring the idea of giving users the ability to search for properties by describing content in photos. In addition this tech can also extract text from images, so street signs, areas, brands etc will become fully searchable.
Another thought is that if the API can tell you what's in the photo, you could dynamically add appropriate ALT tags and naming conventions to all your images to make them easily indexed.
Although it's early days, the technology could prove to be very interesting as it is developed further over time.
This is the same technology that powers the image search in Google Photos and it can recognize anything from flowers, food and animals to local landmarks. Google says it has trained the algorithm to recognize “thousands” of different objects.