CEVA’s CEVA-XM4 is a programmable, digital signal processor and memory subsystem that is offered with a demonstration board for prototyping. The CEVA-XM4 uses scalar and vector units to speed image processing functions. It employs both a very long instruction word (VLIW) and a single instruction multiple data (SIMD) architecture and features a memory sub-system for integration into system-on-chips (SOCs). The processor’s capabilities can support real-time 3D depth map generation and point cloud processing for 3D scanning. The processor can analyze scene information using object detection and recognition algorithms. The architecture features parallel random memory access and is supplied with a hardware development platform with vision libraries and applications including digital video stabilization. The CEVA-XM4 vision processor targets embedded vision applications such as smart phone and tablet applications.
To Learn More:
Headquarters: Mountain View, CA, USA
Product: CEVA-XM4 intelligent vision processor
Key Features: Combines depth generation with vision processor, enables multi-app processing, implements multi-frame algorithms in high resolution.
Share new products that you think are particularly interesting or helpful by contacting James Carroll, Senior Web Editor, Vision Systems Design.