Hong Kong, China, June 28, 2017 /ChinaNewswire.com/ - #8226; CEVA-XM4 imaging and vision DSP enables up to 60x performance improvement for PiSoft’s vision algorithms versus implementation on a CPU
#8226; Rockchip RV1108 visual processors powered by CEVA-XM4 enable distortion correction and stitching on-board for 3K video in real-time
CEVA, Inc. (NASDAQ: CEVA), the leading licensor of signal processing IP for smarter, connected devices, together with Pisoftware Technology Co., Ltd. (“PiSoft”), today announced that PiSoft has developed a complete 360 degree panoramic camera solution for ODMs and OEMs. The solution utilizes the Rockchip RV1108 visual processor which leverages the CEVA-XM4 imaging and vision DSP to run advanced imaging and computer vision algorithms, including real-time distortion correction and panoramic stitching. The solution has already been adopted by ODMs and products are available in the market.
PiSoft developed their solution to address the growing demand for 360 degree cameras, which are being rapidly adopted, in part to generate more virtual reality content. Utilizing their capabilities as a complete technology solution provider, PiSoft built a system on the RV1108 processor, that allows them fully exploit their vision algorithm expertize via the CEVA-XM4 engine and generate outstanding 360 degree image and video quality. Compared to implementing their algorithms on the CPU within the system, the CEVA-XM4 was 36x more efficient for distortion correction and 60x more efficient for stitching. These considerable performance advantages together with the low power performance, allowed PiSoft to fully exploit the capabilities of their software, and create a product, which offers a unique and differentiated experience from competing solutions, while consuming significantly less power.
“To bring our advanced vision expertize to the market, we chose the RV1108 chip with the CEVA-XM4 DSP due to the outstanding performance we could achieve for running our complex vision algorithms,” said Adler Shen, CEO of PiSoft. “CEVA’s powerful suite of development tools ensured we easily and efficiently developed our vision algorithms and allowed us to meet our development goals and reach production ODM customers in record time.”
“We applaud PiSoft’s ability to develop a full solution for panoramic cameras, from system design to software implementation, that offers exceptional 360 degree video and image quality,” said Ilan Yona, vice president and general manager of the Vision Business Unit at CEVA. “Our CEVA-XM4 DSP and software development environment, plays an essential role in the success of this product, enabling PiSoft achieve a huge performance advantage over alternative systems that rely on the CPU or GPU to run vision algorithms.”
RV1108 is a high performance low power application processor. It is embedded with a CEVA-XM4 DSP for imaging and computer vision algorithms and an ARM Cortex-A7 single core processor for system and application. It is a highly-integrated and cost-efficient SoC, which can support H.264 video encoder/decoder up to 1440p, up to 4 camera inputs simultaneously, merge different camera sources together and display on one screen. It is designed for various application scenarios such as car DVR, sports DV, secure camera and UAV camera. For more information, visit
CEVA’s latest generation imaging and vision DSP platforms address the extreme processing requirements and low power constraints of the most sophisticated machine learning and machine vision applications used in smartphones, surveillance, augmented reality, Sense and Avoid drones and self-driving cars. These DSP-based platforms include a hybrid architecture composed of scalar and vector DSP processors coupled with a comprehensive Application Development Kit (ADK) to streamline software deployment. The CEVA ADK includes: CEVA-Link for seamless software level integration with a host processor; a range of widely used and optimized software algorithms; the CEVA Deep Neural Network (CDNN2) real-time Neural Network software framework which streamlines machine learning deployment at a fraction of the power consumption of the leading GPU-based systems, and; state-of-the-art development and debugging tools. For more information, visit http://www.ceva-dsp.com/app/imaging-computer-vision/.
About Pisoftware Technology Co., Ltd
Pisoftware Technology Co., Ltd. (or “PiSoft”) is a tech-based startup in Shenzhen offering VR panoramic image capturing-transmitting-playing solutions including image capturing, content transmission and content distribution to customers in different fields. PiSoft is currently the only tech-based company in China that can mass-produce VR (panoramic+3D) live/recorded broadcasting devices equipped with 4K multi-channel cameras and possesses a series of world-leading core algorithms such as multi-camera real-time fusion algorithm, panoramic+3D algorithm, panoramic audio-video synchronization and particularly important the industrial level automatic calibration technology that can auto-adapt to complex optical lens sets. For more information, visit http://www.pisofttech.com.
About CEVA, Inc.
CEVA is the leading licensor of signal processing IP for a smarter, connected world. We partner with semiconductor companies and OEMs worldwide to create power-efficient, intelligent and connected devices for a range of end markets, including mobile, consumer, automotive, industrial and IoT. Our ultra-low-power IPs for vision, audio, communications and connectivity include comprehensive DSP-based platforms for LTE/LTE-A/5G baseband processing in handsets, infrastructure and machine-to-machine devices, advanced imaging, computer vision and deep learning for any camera-enabled device, audio/voice/speech and ultra-low power always-on/sensing applications for multiple IoT markets. For connectivity, we offer the industry’s most widely adopted IPs for Bluetooth (low energy and dual mode), Wi-Fi (802.11 a/b/g/n/ac up to 4×4) and serial storage (SATA and SAS). Visit us at www.ceva-dsp.com and follow us on Twitter, YouTube and LinkedIn.Source: ChinaNewswire.com
Visit the original: