http://rdf.ncbi.nlm.nih.gov/pubchem/patent/CN-106649542-B
Outgoing Links
Predicate | Object |
---|---|
classificationCPCAdditional | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-044 |
classificationCPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-084 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-06 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-214 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-045 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F40-30 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06T1-60 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-2431 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N5-04 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F18-24 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-90332 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-532 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06N3-02 http://rdf.ncbi.nlm.nih.gov/pubchem/patentcpc/G06F16-583 |
classificationIPCInventive | http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06K9-62 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-583 http://rdf.ncbi.nlm.nih.gov/pubchem/patentipc/G06F16-9032 |
filingDate | 2016-11-02-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
grantDate | 2020-06-16-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationDate | 2020-06-16-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
publicationNumber | CN-106649542-B |
titleOfInvention | System and method for visual question answering |
abstract | Described herein are systems and methods for generating and using an attention-based deep learning architecture for visual question answering (VQA) to automatically generate answers to image (still or video) related questions. To generate correct answers, it is important that the model's attention focus on relevant regions in the image according to the question, since different questions may ask about the properties of different image regions. In an embodiment, such question-guided attention is learned using a configurable convolutional neural network (ABC-CNN). The implementation of the ABC‑CNN model determines the attention map by convolving the image feature map with a configurable convolution kernel determined by question semantics. In an embodiment, the question-guided attention map focuses on areas relevant to the question, and filters out noise in irrelevant areas. |
priorityDate | 2015-11-03-04:00^^<http://www.w3.org/2001/XMLSchema#date> |
type | http://data.epo.org/linked-data/def/patent/Publication |
Incoming Links
Total number of triples: 33.