http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-113269808-A
Outgoing Links
| Predicate | Object |
|---|---|
| assignee | http://rdf.ncbi.nlm.nih.gov/pubchem/patentassignee/MD5_217f2be73fd6be0be36dd55803d3bec5 |
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10016 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20084 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 |
| classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-269 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-246 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-08 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-246 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-269 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 |
| filingDate | 2021-04-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| inventor | http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_81edef06d13925d126cdb9cf2b60b481 http://rdf.ncbi.nlm.nih.gov/pubchem/patentinventor/MD5_fa3d422cf1dc223ca24725b6f92de769 |
| publicationDate | 2021-08-17-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-113269808-A |
| titleOfInvention | Video small target tracking method and device |
| abstract | The small target tracking method and device proposed by the present invention includes a model training stage, a target tracking stage and a model updating stage. In the model training stage, the parameters of the convolutional neural network in the entire tracking model including the self-attention module are determined; in the tracking stage, the target position is continuously detected according to the trained model; The parameters of different modules of the tracking model are updated to ensure continuous accurate and robust tracking effect. The invention combines the multiple features of the moving target to perform the target tracking process, has higher anti-interference ability and robustness, and obtains the weight map corresponding to each feature response map through the self-attention module constructed by the convolutional neural network, Extending the traditional single fusion coefficient into a two-dimensional fusion coefficient matrix (called "attention map") consistent with the size of the response map makes the feature fusion more accurate and has stronger adaptability to tracking in different scenarios. |
| isCitedBy | http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-114245206-A |
| priorityDate | 2021-04-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 25.