Smart imaging for power-efficient extraction of Viola-Jones local descriptors

Abstract

In computer vision, local descriptors permit to summarize relevant visual cues through feature vectors. These vectors constitute inputs for trained classifiers which in turn enable different high-level vision tasks. While local descriptors certainly alleviate the computation load of subsequent processing stages by preventing them from handling raw images, they still have to deal with individual pixels. Feature vector extraction can thus become a major limitation for conventional embedded vision hardware. In this paper, we present a power-efficient sensingprocessing array conceived to provide the computation of integral images at different scales. These images are intermediate representations that speed up feature extraction. In particular, the mixed-signal array operation is tailored for extraction of Haar-like features. These features feed the cascade of classifiers at the core of the Viola-Jones framework. The processing lattice has been designed for the standard UMC 0.18µm 1P6M CMOS process. In addition to integral image computation, the array can be reprogrammed to deliver other early vision tasks: concurrent rectangular area sum, block-wise HDR imaging, Gaussian pyramids and image pre-warping for subsequent reduced kernel filtering.

Multi-Camera Active-Vision for Markerless Shape Recovery of Unknown Deforming Objects

“…(1) they are the most stable sparse features for tracking implementations [56], [166]- [168] and (2) shown to operate faster than SIFT at the cost of matching stability [56].…”

A 26.5 nJ/px 2.64 Mpx/s CMOS vision sensor for Gaussian pyramid extraction

“…Besides this, analysis show that parallel-processing vision architectures are largely tolerant to individual processor errors, e.g. deviations close to 10% are tolerated in many cases [2].…”