Accentuating Features of Description Logics in High-Level Interpretations of Hand-Drawn Sketches
Zusammenfassung
We propose an ontology-based approach to interpret hand-drawn sketches, originating from empirical results of experiments with human participants. The approach combines qualitative features of the sequence of sketch strokes with a high-level knowledge, and accentuates the potential effectiveness of interpretation via description logics. The results of an implementation, along with explanations, are presented to show how to extract the semantics of hand-drawn sketches of four object categories.
- Vollständige Referenz
- BibTeX
Abdelghaffar, N. M., Abdelfattah, A. M., Taha, A. A. & Khamis, S. M.,
(2019).
Accentuating Features of Description Logics in High-Level Interpretations of Hand-Drawn Sketches.
KI - Künstliche Intelligenz: Vol. 33, No. 3.
Springer.
(S. 253-265).
DOI: 10.1007/s13218-019-00602-4
@article{mci/Abdelghaffar2019,
author = {Abdelghaffar, Nashwa M. AND Abdelfattah, Ahmed M. H. AND Taha, Azza A. AND Khamis, Soheir M.},
title = {Accentuating Features of Description Logics in High-Level Interpretations of Hand-Drawn Sketches},
journal = {KI - Künstliche Intelligenz},
volume = {33},
number = {3},
year = {2019},
,
pages = { 253-265 } ,
doi = { 10.1007/s13218-019-00602-4 }
}
author = {Abdelghaffar, Nashwa M. AND Abdelfattah, Ahmed M. H. AND Taha, Azza A. AND Khamis, Soheir M.},
title = {Accentuating Features of Description Logics in High-Level Interpretations of Hand-Drawn Sketches},
journal = {KI - Künstliche Intelligenz},
volume = {33},
number = {3},
year = {2019},
,
pages = { 253-265 } ,
doi = { 10.1007/s13218-019-00602-4 }
}
Sollte hier kein Volltext (PDF) verlinkt sein, dann kann es sein, dass dieser aus verschiedenen Gruenden (z.B. Lizenzen oder Copyright) nur in einer anderen Digital Library verfuegbar ist. Versuchen Sie in diesem Fall einen Zugriff ueber die verlinkte DOI: 10.1007/s13218-019-00602-4
Haben Sie fehlerhafte Angaben entdeckt? Sagen Sie uns Bescheid: Feedback abschicken
Mehr Information
ISSN: 1610-1987
Datum: 2019
Typ: Text/Journal Article

