VeriSilicon's Vivante DeWarp Processor provides high-performance DeWarp processing for the correction of the distortion that is introduced in images produced by fisheye and wide-angle lenses. It is implemented with a line/tile-cache based architecture. With configurable address mapping look up tables and per tile processing, it can successfully generate corrected output images.
This DeWarp IP is designed for easy integration into SoCs, providing high performance, high quality, low power consumption, and the smallest silicon footprint for its class. Its core is delivered as synthesizable RTL. DeWarp IP is technology independent, which can be synthesized using a variety of libraries. Dynamic power consumption is minimized by extensive use of multi-level hierarchical clock gating.
DeWarp Processor IP
Overview
Key Features
- DeWarp Engine Features
- Fisheye correction
- Wide Field of View (FOV) correction
- Keystone correction
- Scaling up from 1x to 4x, apply in DeWarp images
- Fisheye 180° /360° view
- 4PTZ view
- Mirror, Flip
- Stream interface output
- Video Scaling Engine Features
- Scale Engine, arbitrary scale ratio
- Pyramid Engine, configurable scale instances
- Stream interface input and output
- DMA read from memory
- Option add-on Features
- 10bit processing
- Grey Image support
Technical Specifications
Foundry, Node
All
Maturity
Mass Production
Availability
now
Related IPs
- AI inference processor IP
- Graphics Processor Overlay IP Core
- RISC-V ARC-V RMX-100 Ultra-low Power 32-bit Processor IP for Embedded Applications
- DES Encryption and Decryption Processor
- Embedded Configuration and Test Processor
- 32-bit RISC Processor To Deliver High Performance In Low-Cost Microcontroller Applications