Statements (29)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:facial_recognition_system
|
| gptkbp:accuracy |
97.35%
|
| gptkbp:application |
face identification
face verification |
| gptkbp:architecture |
nine-layer deep neural network
|
| gptkbp:author |
gptkb:Marc'Aurelio_Ranzato
gptkb:Ming_Yang Lior Wolf Yaniv Taigman |
| gptkbp:comparesTo |
human-level performance
|
| gptkbp:developedBy |
gptkb:Facebook
|
| gptkbp:input |
face images
|
| gptkbp:introducedIn |
2014
|
| gptkbp:language |
gptkb:Python
|
| gptkbp:notableFor |
closing the gap to human-level face verification accuracy
|
| gptkbp:notablePublication |
DeepFace: Closing the Gap to Human-Level Performance in Face Verification
|
| gptkbp:output |
face representations
|
| gptkbp:publicationDate |
2014 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
|
| gptkbp:relatedTo |
gptkb:artificial_intelligence
computer vision face recognition |
| gptkbp:trainer |
4 million facial images
4,000 identities |
| gptkbp:uses |
convolutional neural networks
deep learning 3D face alignment |
| gptkbp:bfsParent |
gptkb:Kalman_Taigman
|
| gptkbp:bfsLayer |
8
|
| https://www.w3.org/2000/01/rdf-schema#label |
DeepFace
|