.png){kind=icon}
.png){kind=icon}
Return
Comic character recognition using graph embedding in graph representations
.png){kind=icon}
PRE
AI
DOI：10.1007/s10032-025-00568-0.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Comic books are becoming a significant form of entertainment, instruction, and advertising on a global scale. Comics frequently use text balloons and illustrations to communicate stories. The identification of comic characters has grown to be a particularly fascinating subject of study in the field of comic analysis. For digital comic books, it enables incredibly effective indexation and comic character retrieval. For the recognition, retrieval, and indexation of comic characters, several approaches have been developed. In this research, we suggested a graph embedding approach for comic character recognition. We suggested a novel method for character recognition using graphs. Modern comic character identification techniques are assessed with our solution using standard graph-based datasets for testing. The base method presented by authors of the dataset is outperformed by our suggested methodology. The suggested method may be viewed as a solid foundational method for identifying comic book characters.
Keywords:
Comic recognition
Graph neural networks
Graph embedding
Document analysis
Comic book analysis
Deep learning
Journal
I
IF:
2.5
Papers:
36
Citations:
826
View more organizations