论文信息 - Sketched Symbol Recognition with a Latent-Dynamic Conditional Model

Sketched Symbol Recognition with a Latent-Dynamic Conditional Model

In this paper we present a recognizer of sketched symbols based on Latent-Dynamic Conditional Random Fields (LDCRF), a discriminative model for sequence classification. The LDCRF model classifies unsegmented sequences of strokes into domain symbols by taking into account contextual and temporal information. In particular, LDCRFs learn the extrinsic dynamics among strokes by modeling a continuous stream of symbol labels, and learn internal stroke sub-structure by using intermediate hidden states. The performance of our work is evaluated in the electric circuit domain.

Michele Risi | Genny Tortora | Vincenzo Deufemia

[1] Benjamin King. Step-Wise Clustering Procedures , 1967 .

[2] Andrew McCallum,et al. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[3] Thomas F. Stahovich,et al. Generating Multiple New Designs from a Sketch , 1996, Artif. Intell..

[4] Randall Davis,et al. Tahuti: a geometrical sketch recognition system for UML class diagrams , 2006, SIGGRAPH Courses.

[5] Michael McGill,et al. Introduction to Modern Information Retrieval , 1983 .

[6] Trevor Darrell,et al. Latent-Dynamic Discriminative Models for Continuous Gesture Recognition , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.