Motion-vector estimation and cognitive classification on an image sensor/processor 3D stacked system featuring ThruChip interfaces

Abstract

1,000 fps motion vector estimation and classification engine for highspeed computational imaging in a 3D stacked imager/processor module is proposed, prototyped, assembled, and also tested. The module features ThruChip interfaces for high fps image transfer, orders of magnitude more area/power efficient motion vector estimation architecture compared to conventional ones, and a cognitive classification scheme employed on motion vector patterns, enabling the classification of moving objects not possible in conventional proposals.