BRIEF: Computing a Local Binary Descriptor Very Fast

TLDR

Binary descriptors are increasingly popular for fast, memory‑efficient feature comparison, typically produced by first computing floating‑point descriptors such as SIFT and then binarizing them. This paper proposes BRIEF, a binary descriptor computed directly from simple intensity difference tests. BRIEF is computed directly from simple intensity difference tests without first generating floating‑point descriptors. BRIEF is extremely fast to build and match, achieving recognition accuracy comparable to SURF and SIFT while running in a fraction of their time.

Abstract

Binary descriptors are becoming increasingly popular as a means to compare feature points very fast while requiring comparatively small amounts of memory. The typical approach to creating them is to first compute floating-point ones, using an algorithm such as SIFT, and then to binarize them. In this paper, we show that we can directly compute a binary descriptor, which we call BRIEF, on the basis of simple intensity difference tests. As a result, BRIEF is very fast both to build and to match. We compare it against SURF and SIFT on standard benchmarks and show that it yields comparable recognition accuracy, while running in an almost vanishing fraction of the time required by either.