http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113139479-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-241 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V10-44 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06V40-174 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-764 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-44 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V40-16 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06V10-82 |
filingDate | 2021-04-28-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2022-07-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2022-07-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-113139479-B |
titleOfInvention | A micro-expression recognition method and system based on optical flow and RGB modal contrast learning |
abstract | The invention relates to a micro-expression recognition method and system based on optical flow and RGB modal contrastive learning, including: A. Preprocessing micro-expression videos, including: acquiring video frame sequences, face detection and positioning, and face alignment And extract optical flow sequence features and RGB sequence features; B. Extract optical flow sequence features and RGB sequence features from the micro-expression dataset, and divide it into a test set and a training set; C. Build a dual-modal contrastive learning and recognition model, including Three-dimensional convolutional residual network, three-dimensional convolutional residual network encodes optical flow sequence features and RGB sequence features respectively; D. Construct cross entropy loss and contrastive learning loss, and use loss function to train dual-modal contrastive learning and recognition model; E. Classification and recognition: According to the trained dual-modal contrast learning recognition model, the test set is classified and recognized. The present invention uses supervised information and unsupervised information to constrain the network at the same time, so as to obtain stronger feature expression. |
priorityDate | 2021-04-28-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Predicate | Subject |
---|---|
isDiscussedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID426109005 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID6442267 |
Total number of triples: 22.