http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-110910336-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T2207-20221 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T7-55 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T5-50 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06N3-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T7-55 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06T5-50 |
filingDate | 2019-10-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2022-08-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2022-08-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-110910336-B |
titleOfInvention | A Stereo High Dynamic Range Imaging Method Based on Fully Convolutional Neural Networks |
abstract | The invention discloses a stereo high dynamic range imaging method based on a fully convolutional neural network, which uses an exposure calibration network to convert a left-viewpoint underexposure image and a right-viewpoint overexposure image to the same exposure; Perform forward rendering on left-view underexposed images and right-view over-exposed images to generate right-view under-exposed images and left-view over-exposed images; introduce additional exposure information to guide hole filling in the rendered images; utilize HDR image fusion The network extracts fusion features, fuses the left-view over-exposed image after hole filling and the original left-view under-exposure image into a left-view HDR image, and fuses the hole-filled right-view under-exposed image and the original right-view over-exposed image into a right-view HDR image. Viewpoint HDR image; and then obtain a stereoscopic HDR image; the advantage is that the dynamic range of the original stereoscopic LDR image can be improved, and the detail information in the underexposed and overexposed areas of the original stereoscopic LDR image can be reconstructed. |
priorityDate | 2019-10-30-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 18.