http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110503661-A
Outgoing Links
Predicate | Object |
---|---|
assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_c16523724d9aa50f82e1a3bb35e8c365 |
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20084 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10016 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-251 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-246 |
filingDate | 2018-05-16-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_4ed399afcbe9ccadb33853cdf3568033 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_7910e00d8f0867ecfac66f9cd7375975 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_c1086ff74ed2e13a9efc381af89c8214 |
publicationDate | 2019-11-26-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-110503661-A |
titleOfInvention | A Target Image Tracking Method Based on Deep Reinforcement Learning and Spatiotemporal Context |
abstract | The invention discloses a target image tracking method based on deep reinforcement learning and spatio-temporal context, comprising the following steps: 1) at each time step t, using a feature extraction network to obtain an image x t from an input sequence as a visual feature; The visual features pass through the STC and the recurrent neural network, and then extract the spatio-temporal feature c t and hidden layer state h t from the STC and the recurrent neural network respectively, where the spatio-temporal feature c t will be used as the reference standard; 2) Model building; 3) Model training ; 4) Carry out target tracking according to the predicted position of the model. The method and model proposed by the present invention have a high success rate and accuracy score in the tracking process, which also reflects that the DRST model based on reinforcement learning and spatio-temporal context proposed by the present invention can realize long-term tracking of the target object and avoid the occurrence of Track drift. |
isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113034378-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111862158-A http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111862158-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111539979-B http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111539979-A |
priorityDate | 2018-05-16-04:00^^<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/SID419479910 http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID5709 |
Total number of triples: 25.