Visual Relationship Detection
GPTKB entity
Statements (28)
Predicate | Object |
---|---|
gptkbp:instanceOf |
computer vision task
|
gptkbp:application |
autonomous vehicles
robotics image captioning visual question answering |
gptkbp:bench |
gptkb:Open_Images
gptkb:Visual_Genome VRD dataset |
gptkbp:challenge |
ambiguity in relationships
data annotation complexity long-tail distribution of relationships |
gptkbp:field |
gptkb:artificial_intelligence
computer vision |
gptkbp:goal |
detect relationships between objects in images
|
https://www.w3.org/2000/01/rdf-schema#label |
Visual Relationship Detection
|
gptkbp:input |
gptkb:illustrator
|
gptkbp:notablePublication |
Visual Relationship Detection with Language Priors (Lu et al., 2016)
|
gptkbp:output |
triples of (subject, predicate, object)
|
gptkbp:proposedBy |
2016
|
gptkbp:relatedTo |
object detection
scene understanding image understanding |
gptkbp:uses |
convolutional neural networks
deep learning graph neural networks |
gptkbp:bfsParent |
gptkb:VRD
gptkb:Open_Images_Challenge |
gptkbp:bfsLayer |
7
|