Efficient inner product computation for image and video analysis

摘要

A computationally efficient approach to determining inner products between feature vectors is provided that eliminates or reduces the need for multiplication, and more specifically, provides an efficient and accurate basis selection for techniques such as Orthogonal Matching Pursuit.

主权项

1. A method for machine-implemented image feature processing comprising:
accepting a data representation of a plurality of m dimensional feature vectors ν representing a processing of an image or video signal; for each feature vector ν, forming a p dimensional binary vector h from the data representation of the feature vector ν using a first procedure that includes
transforming the feature vector ν to form a transformed feature vector {tilde over (ν)}, including applying a machine implemented computation to compute elements of the transformed feature vector as an additive combination of elements of the feature vector, anddetermining each element of h according to a sign of a selected additive combination of elements of {tilde over (ν)}; for each feature vector ν forming a data transformation of the feature vector ν including computing at least one approximation of an inner product of ν with another vector u using a second procedure including comparing the binary vector h formed from ν with a binary vector formed from u; performing an image analysis process according to the plurality of transformed feature vectors.