http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-111429436-B
Outgoing Links
| Predicate | Object |
|---|---|
| classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-30168 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20081 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-10024 |
| classificationCPCInventive | 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/G06T7-0002 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-044 |
| classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-08 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-00 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 |
| filingDate | 2020-03-29-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| grantDate | 2022-03-15-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationDate | 2022-03-15-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
| publicationNumber | CN-111429436-B |
| titleOfInvention | An essential image analysis method based on multi-scale attention and label loss |
| abstract | The present invention proposes an essential image analysis method based on multi-scale attention and label loss, introduces the circular convolution attention mechanism and confrontation idea into the essential decomposition problem, and constructs a multi-scale attention MSA- for essential image analysis. Net network, the network structure follows the basic framework of Generative Adversarial Network (GAN), including generator and discriminator. The generator consists of two parts, an attention sub-network and an encoder-decoder sub-network, which are used to decompose the image into reflection maps and light maps. The role of the discriminator is to give the probability that the image is the correct essential image for any input image. At the same time, the invention also provides a new label loss function for improving the reflection map decomposition effect. The loss function is constructed based on the label image (ground truth) in the data set, so that the reflection map obtained by network decomposition has better local Texture consistency effect and quantitative evaluation index. |
| priorityDate | 2020-03-29-04:00^^<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/compound/CID31373 http://rdf.ncbi.nlm.nih.gov/pubchem/substance/SID419556474 |
Total number of triples: 20.