US 9,811,540 B2
Compact, clustering-based indexes for large-scale real-time lookups on streaming videos
Min Feng, Princeton, NJ (US); Giuseppe Coviello, Plainsboro, NJ (US); Srimat Chakradhar, Manalapan, NJ (US); Nitin Agrawal, East Brunswick, NJ (US); and Yi Yang, Plainsboro, NJ (US)
Assigned to NEC Corporation, Tokyo (JP)
Filed by NEC Laboratories America, Inc., Princeton, NJ (US)
Filed on Apr. 1, 2016, as Appl. No. 15/88,452.
Claims priority of provisional application 62/144,626, filed on Apr. 8, 2015.
Prior Publication US 2016/0299920 A1, Oct. 13, 2016
Int. Cl. G06K 9/00 (2006.01); G06F 17/30 (2006.01); G06K 9/62 (2006.01)
CPC G06F 17/30277 (2013.01) [G06F 17/30256 (2013.01); G06K 9/00268 (2013.01); G06K 9/00711 (2013.01); G06K 9/6219 (2013.01); G06K 9/6223 (2013.01); G06K 9/6272 (2013.01)] 10 Claims
OG exemplary drawing
1. A method for recognizing a face, comprising:
receiving images of training faces;
generating feature vectors of the images;
generating clusters from the feature vectors each with one or more centroids or a cluster representative;
for a query to search for a query face, generating query feature vectors for the query face and comparing the query feature vectors with the centroids of all clusters to find one or more similar clusters;
for clusters above a similarity threshold, comparing feature vectors of corresponding members of the clusters with the query feature vectors; and
indicating as matching candidates for cluster members with similarity above a threshold, wherein each cluster model size is sub-linear or logarithmic in the number of the training faces (or features) in a database.