http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113887473-A

Outgoing Links

Predicate Object
assignee http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_e89c206df73c7928c3fa05f87f1287de
classificationCPCInventive http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08
http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045
classificationIPCInventive http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-00
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08
http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04
filingDate 2021-10-15-04:00^^<http://www.w3.org/2001/XMLSchema#date>
inventor http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_9f4ed342b4dd7f5991b782352f5067a5
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_4d21f36d1930549eaa3e558ae7646fbb
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_092ce9ebea8e97a8fe1cadaa85901747
http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_8eab28bf2baa03c8e4eb8ebc3df660ec
publicationDate 2022-01-04-04:00^^<http://www.w3.org/2001/XMLSchema#date>
publicationNumber CN-113887473-A
titleOfInvention An Improved Normalized Deformable Convolutional Crowd Counting Method
abstract The invention belongs to the technical field of computer vision, and in particular relates to a method for estimating the number of people by using a convolutional neural network. An improved normalized deformable convolutional crowd counting method, comprising: constructing a normalized deformable convolutional neural network; using the normalized deformable convolutional neural network to constrain the position of sampling points of a feature map of an input image to obtain accurate human head characteristics. The present invention proposes a normalized deformable convolution (NDConv), which limits the offset of the sampling points to a certain extent, can obtain the information in the effective area of the sampling points, and does not increase the extra calculation amount, and improves the neural network. The accuracy of the network's prediction of crowd counts.
priorityDate 2021-10-15-04:00^^<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/SID415828022
http://rdf.ncbi.nlm.nih.gov/pubchem/compound/CID130693

Total number of triples: 19.