Fig. 2From: Querying geo-tagged videos for vision applications using spatial metadataThe framework of GIFTBack to article page