By combining this technology with its existing core facial image sensing technology, gesture recognition can be started automatically based on the analysis of interrelation between the position or direction of the face and the hand shape or position. This avoids the need of conventional gesture recognition technologies that require a trigger motion such as waving a hand to start the recognition process.
The technology allows users to control or manipulate devices in a more natural manner, such as operating a TV set with hand movements or releasing the shutter of a camera via a hand gesture. It employs techniques such as statistical classification of a large number of existing images and model-fitting technology, both of which were developed over several years inside the company. Through the modeling of hand shapes, this new technology enables fast recognition of gestures using a small amount of memory. It recognises hand gestures from a an image as small as 40 pixels from very near (about 10-cm) to far field (several meters) and can be used on smartphones and tablets with a recognitiion spee of 30 frame/s using a 1 GHz Qualcomm SnapDragon processor. The small program size means it can be embedded in a wide range of other devices.
OMRON has been involved with facial recognition with its "OKAO Vision" technology since 1995. This facial image sensing technology is capable of detecting various kinds of information from facial images, and is one of OMRON's core technologies. The developments to date have enabled such functions as face detection, face recognition, smile degree estimation, estimation of gaze direction, detection of the degree of mouth or eye opening, and estimation of age and gender. The technology has been used for the auto-focus function of digital cameras and smartphones, as well as the skin beautification function of printers, image management for personal computers and many other applications.
The company says it will further develop its image sensing technology