Lattice Semiconductor Corporation today announced it will demonstrate 3D Impact Media’s (3DIM) RealityBox, a low-density LatticeECP3 field-programmable gate array (FPGA) based real-time three-dimensional (3D) video converter, in its private meeting suite at the International Consumer Electronics Show (CES) on 8-11 January 2013 in the East Tower of the Las Vegas Hotel (suite 2980) in Las Vegas, Nevada (US). To register for a specific time to visit with Lattice, please see this website.
Using the RealityBox, any two-dimensional (2D) or stereoscopic 3D video stream can be converted and shown on glasses-free 3D displays in real-time. This capability enables new applications such as broadcasting of live events in public venues or digital signage, where 3D content is seldom available and expensive.
The RealityBox hardware is designed to support full 1080p resolution up to 60 frames per second (fps), or 720p up to 120fps, using an HDMI interface. The video pipeline supports 2D/S-3D to S-3D/AS-3D conversion, real time playback of 2D+Z videos and allows disparity analysis of S-3D input signals. The main building blocks are an input/output scalar, disparity/motion estimator, temporal stability filter, image peaking, contrast, brightness adjustment and a programmable multi-view generator. The hardware can be used for barrier and lenticular technology-based glasses-free 3D displays. All parameters are freely programmable to support different types of 3D displays, including fractional views. The maximum amount of views that can be generated is 1024.
The 3DIM conversion pipeline has an onboard on-screen display (OSD) that can be customised and includes ticker and image overlay options that can be added to the video stream. Due to its flexible design and low latency, as well as the LatticeECP3 device’s low power, the RealityBox can be used in many different 3D applications, including next generation gaming machines and in-flight entertainment. In medical applications the RealityBox can help improve depth perception, allowing for more precise surgical procedures, and in automotive applications it can be used for object recognition and assistance to make driving safer.