Techniques for cloud-based similarity searches

摘要: Techniques for facilitating a similarity search of digital assets (e.g., audio files, image video etc.) are described. Consistent with some embodiments, cloud-based service manages one or more tree data structures use in organizing to make the searchable. Each asset is associated feature vector based on various attributes and/or characteristics asset. The then assigned leaf nodes measure distance between and virtual node. When similar invoked, prioritized breadth first performed identify having vectors closest reference