25 October 2016 Convolutional neural network for pottery retrieval
Author Affiliations +
J. of Electronic Imaging, 26(1), 011005 (2016). doi:10.1117/1.JEI.26.1.011005
The effectiveness of the convolutional neural network (CNN) has already been demonstrated in many challenging tasks of computer vision, such as image retrieval, action recognition, and object classification. This paper specifically exploits CNN to design local descriptors for content-based retrieval of complete or nearly complete three-dimensional (3-D) vessel replicas. Based on vector quantization, the designed descriptors are clustered to form a shape vocabulary. Then, each 3-D object is associated to a set of clusters (words) in that vocabulary. Finally, a weighted vector counting the occurrences of every word is computed. The reported experimental results on the 3-D pottery benchmark show the superior performance of the proposed method.
© 2016 SPIE and IS&T
Halim Benhabiles, Hedi Tabia, "Convolutional neural network for pottery retrieval," Journal of Electronic Imaging 26(1), 011005 (25 October 2016). https://doi.org/10.1117/1.JEI.26.1.011005


Research on image matching method of big data image of...
Proceedings of SPIE (December 14 2015)
Computer vision research with new imaging technology
Proceedings of SPIE (December 14 2015)
Lossless description of 3D range models
Proceedings of SPIE (February 15 2012)
Relevance feedback-based building recognition
Proceedings of SPIE (July 14 2010)
Histogram as a measure of shape matching
Proceedings of SPIE (May 28 2004)

Back to Top