The IVP is extremely power efficient. As an example, for IVP implemented in an automatic synthesis, P&R flow in 28nm HPM process, regular VT, a 32-bit integral image computation on 16b pixel data at 1080p30 consumes 10.8 mW. The integral image function is commonly used in applications such as face and object detection and gesture recognition.
Tensilica also announced a number of alliances with software vendors Dreamchip, Almalence, Irida Labs, and Morpho.
|The IVP Core Architecture|
With Sample Memory Sizes Selected
EETimes talks about a battle between Tensilica and Ceva imaging IP cores. No clear winnder is declared, but Tensilica IVP is said to have more processing power. Another EETimes article talks about ISP IP applications and requirements.