Bag of Dense Derivative Depth Patterns (bag_d3p)

Description

Bag-D3P is a Matlab-based software specially designed for biometric face recognition using depth cameras as input information. The Bag-D3P descriptor is composed of four different stages that fully exploit the characteristics of depth information: 1) dense spatial derivatives to encode the 3-D local structure; 2) face-adaptive quantization of the previous derivatives; 3) multibag of words that creates a compact vector description from the quantized derivatives; and 4) spatial block division to add global spatial information. The proposed system can recognize people faces from a wide range of poses, not only frontal ones, increasing its applicability to real situations. Last, a new face database of high-resolution depth images has been created and made it public for evaluation purposes.

Code

• Link
• Github: https://github.com/gti-upm/bag_d3p

Database

Results that appear on the related paper are obtained over the HRRFaceD dataset that is publicly available

Paper

T. Mantecón, C.R. del Blanco, F. Jaureguizar, N. García, “Visual Face Recognition using Bag of Dense Derivative Depth Patterns“, IEEE Signal Processing Letters, vol. 23, no. 6, pp. 771-775, June 2016. (doi: 10.1109/LSP.2016.2553784) [Link]

Contact

Tomás Mantecón: http://gti.ssr.upm.es/tomas-mantecon

Carlos R. del Blanco Adán: http://gti.ssr.upm.es/carlos-r-del-blanco