An image processing system identifies desired views of a particular subject and triggers a mobile camera device in proximity to the subject to capture a digital image, thus collecting new image data to augment an existing database. In an exemplary method, location data for a camera-equipped mobile communication device in visual proximity to a subject of interest is received at an image collection server. Upon determining whether the location data corresponds to a desired view of the subject of interest, the image collection server sends an image capture request to the mobile device, triggering the user of the device to capture an image of the subject. Finally, digital image data is received from the mobile communication device in response to the image capture request. The digital image data may then be used for assembling a composite representation of the subject of interest, such as a 3D model of the subject.