GbR is a biennial workshop organized by the IAPR Technical Committee 15, which encourages research works in Pattern Recognition and Image Analysis within the Graph Theory framework. GbR 2025 will provide a forum for presenting and discussing original research results on theoretical aspects and their applications in these domains. This edition encourages works on the more general context of Deep Learning on graphs. The scope of GbR2025 includes, but is not limited to, the following topics :

▪ Graph-based methodologies for pattern recognition
▪ Deep learning on graphs: Graph Neural Networks, Graph pooling, Graph autoencoders, ...
▪ Graph-based learning and clustering
▪ Graph matching and classification
▪ Graph distance and similarity measures
▪ Graph kernels and graph embedding
▪ Graph Signal Processing
▪ Graph based image segmentation
▪ Irregular (graph) pyramids
▪ Graph representation of images and shapes
▪ Graphs in bioinformatics
▪ Data mining on/with graphs
▪ Graphs in social network analysis
▪ Other graph-based applications in pattern recognition

GBR2025 will be held in Caen (Normandy, France), at ENSICAEN on the University Campus.

The workshop proceedings will be published by Springer in the Lecture Notes in Computer Science series.