Relevance feedback techniques are expected to play an important role in 3D search engines, as they help to bridge the semantic gap between the user and the system. Indeed, similarity is a cognitive process that depends on the observer. We propose a novel relevance feedback technique, which relies on the assumption that similarity may emerge from the inhibition of differences, i.e., from the lack of diversity with respect to the shape properties taken into account. To this end, a user is provided with a variety of shape descriptors, each analyzing different shape properties. Then the user expresses his/her multilevel relevance judgements, which correspond to his/her concept of similarity among the retrieved objects. Finally, the system inhibits the role of the shape properties that do not reflect the user’s idea of similarity. The feedback technique is based on a simple scaling procedure, which does not require neither a priori learning nor parameter optimization. We show examples and experiments on a benchmark dataset of 3D models.

3D relevance feedback via multilevel relevance judgements

FROSINI, PATRIZIO;
2010

Abstract

Relevance feedback techniques are expected to play an important role in 3D search engines, as they help to bridge the semantic gap between the user and the system. Indeed, similarity is a cognitive process that depends on the observer. We propose a novel relevance feedback technique, which relies on the assumption that similarity may emerge from the inhibition of differences, i.e., from the lack of diversity with respect to the shape properties taken into account. To this end, a user is provided with a variety of shape descriptors, each analyzing different shape properties. Then the user expresses his/her multilevel relevance judgements, which correspond to his/her concept of similarity among the retrieved objects. Finally, the system inhibits the role of the shape properties that do not reflect the user’s idea of similarity. The feedback technique is based on a simple scaling procedure, which does not require neither a priori learning nor parameter optimization. We show examples and experiments on a benchmark dataset of 3D models.
D. Giorgi; P. Frosini; M. Spagnuolo; B. Falcidieno
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: http://hdl.handle.net/11585/93546
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 18
  • ???jsp.display-item.citation.isi??? 12
social impact