Co-occurrence flow for pedestrian detection

The last few years have seen considerable progress in pedestrian detection. Recent work has established a combination of oriented gradients and optic flow as effective features although the detection rates are still unsatisfactory for practical use. This paper introduces a new type of motion feature, the co-occurrence flow (CoF). The advance is to capture relative movements of different parts of the entire body, unlike existing motion features which extract internal motion in a local fashion. Through evaluations on the TUD-Brussels pedestrian dataset, we show that our motion feature based on co-occurrence flow contributes to boost the performance of existing methods. © 2011 IEEE.

MCBoost: Multiple classifier boosting for perceptual co-clustering of images and visual features

We present a new co-clustering problem of images and visual features. The problem involves a set of non-object images in addition to a set of object images and features to be co-clustered. Co-clustering is performed in a way that maximises discrimination of object images from non-object images, thus emphasizing discriminative features. This provides a way of obtaining perceptual joint-clusters of object images and features. We tackle the problem by simultaneously boosting multiple strong classifiers which compete for images by their expertise. Each boosting classifier is an aggregation of weak-learners, i.e. simple visual features. The obtained classifiers are useful for object detection tasks which exhibit multimodalities, e.g. multi-category and multi-view object detection tasks. Experiments on a set of pedestrian images and a face data set demonstrate that the method yields intuitive image clusters with associated features and is much superior to conventional boosting classifiers in object detection tasks.

Graphene SOI CMOS sensors for detection of PPB levels of NO2 in air

There is considerable demand for sensors that are capable of detecting ultra-low concentrations (sub-PPM) of toxic gases in air. Of particular interest are NO2 and CO that are exhaust products of internal combustion engines. Electrochemical (EC) sensors are widely used to detect these gases and offer the advantages of low power, good selectivity and temporal stability. However, EC sensors are large (1 cm3), hand-made and thus expensive ($25). Consequently, they are unsuitable for the low-cost automotive market that demands units for less than $10. One alternative technology is SnO2 or WO3 resistive gas sensors that are fabricated in volume today using screen-printed films on alumina substrates and operate at 400°C. Unfortunately, they suffer from several disadvantages: power consumption is high 200 mW; reproducibility of the sensing element is poor; and cross-sensitivity is high. © 2013 IEEE.