Combining shape and color clues can improve 3D model retrieval performances. We have introduced an extended version of the spin-image descriptor that can account also for color data. We also uesd multi-branch Convolutional Neural Networks (CNN) for 3D shape retrieval exploiting also surface and volumetric clues.
Recent publications:
Material identification using RF sensors and convolutional neural networks Proceedings Article
In: ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 3662–3666, IEEE 2019.
Deep Learning for 3D Shape Classification based on Volumetric Density and Surface Approximation Clues. Proceedings Article
In: VISIGRAPP (5: VISAPP), pp. 317–324, 2018.
Deep learning for 3D shape classification from multiple depth maps Proceedings Article
In: 2017 IEEE International Conference on Image Processing (ICIP), pp. 3615–3619, IEEE 2017.
3D reconstruction from web harvested images using a forensic quality metric Proceedings Article
In: 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 1997–2001, IEEE 2017.
Combining color and shape descriptors for 3D model retrieval Journal Article
In: Signal Processing: Image Communication, vol. 28, no. 6, pp. 608–623, 2013.