Fast recursive ensemble convolution of haar-like features

Daniel Wesierski (1) , Maher Mkhinini (1) , Patrick Horain (1) , Anna Jezierska (2)

1 TSP - EPH - Département Electronique et Physique
2 LIGM - Laboratoire d'Informatique Gaspard-Monge

Haar-like features are ubiquitous in computer vision, e.g. for Viola and Jones face detection or local descriptors such as Speeded-Up-Robust-Features. They are classically computed in one pass over integral image by reading the values at the feature corners. Here we present a new, general, parsing formalism for convolving them more efficiently. Our method is fully automatic and applicable to an arbitrary set of Haar-like features. The parser reduces the number of memory accesses which are the main computational bottleneck during convolution on modern computer architectures. It first splits the features into simpler kernels. Then it aligns and reuses them where applicable forming an ensemble of recursive convolution trees, which can be computed faster. This is illustrated with experiments, which show a significant speed-up over the classic approach.

Mots clés: Computer architecture; Convolution; Face detection; Feature extraction; Kernel; Vectors

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Daniel Wesierski, Maher Mkhinini, Patrick Horain, Anna Jezierska. Fast recursive ensemble convolution of haar-like features. CVPR "12 : IEEE Conference on Computer Vision and Pattern Recognition, Jun 2012, Providence, Rhode Island, United States. pp.3689-3696, ⟨10.1109/CVPR.2012.6248115⟩. ⟨hal-00819740⟩