US 12,266,174 B2
Few-shot action recognition
Biplob Debnath, Princeton, NJ (US); Srimat Chakradhar, Manalapan, NJ (US); Oliver Po, San Jose, CA (US); Asim Kadav, Mountain View, CA (US); Farley Lai, Santa Clara, CA (US); and Farhan Asif Chowdhury, Albuquerque, NM (US)
Assigned to NEC Corporation, Tokyo (JP)
Filed by NEC Laboratories America, Inc., Princeton, NJ (US)
Filed on Jul. 12, 2022, as Appl. No. 17/862,667.
Claims priority of provisional application 63/220,623, filed on Jul. 12, 2021.
Prior Publication US 2023/0008303 A1, Jan. 12, 2023
Prior Publication US 2023/0049770 A1, Feb. 16, 2023
Int. Cl. G06V 20/40 (2022.01); G06N 3/08 (2023.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01)
CPC G06V 20/41 (2022.01) [G06N 3/08 (2013.01); G06V 10/764 (2022.01); G06V 10/82 (2022.01); G06V 20/46 (2022.01)] 14 Claims
OG exemplary drawing
 
1. A computer-implemented method of training a neural network, comprising:
training a feature extractor and a classifier, the feature extractor including a time segmentation network, using a first set of training data that includes one or more base cases, wherein the training data includes video and the time segmentation network extracts features from a plurality of frames from the video and uses segmental consensus to generate a feature vector corresponding to the plurality of frames; and
training the classifier with few-shot adaptation using a second set of training data, smaller than the first set of training data, while keeping parameters of the feature extractor constant.