LG's high-performance 3D camera module leverages the Rockchip RK1608 with multi-core CEVA-XM4 vision DSP.
CEVA-XM4 vision DSP platform provides advanced capabilities including SLAM, stereo disparity, and 3D point cloud.

CEVA announced a partnership with LG Electronics to provide its solutions for its smart 3D cameras.

The 3D camera module comprises a Rockchip RK1608 coprocessor with multiple CEVA-XM4 imaging and vision DSPs. This enables a variety of 3D sensing applications, including facial recognition biometric authentication, 3D reconstruction technology, gesture recognition, obstacle detection, and AR and VR capabilities.

Computer vision experts from CEVA, Rockchip and LG Electronics have collaborated to implement LG's proprietary algorithms on the CEVA-XM4 using CEVA's software tool kit and optimized algorithm libraries, as well as to ensure optimal performance in constrained power environments.

“Cost-effective 3D camera sensor modules are essential to deliver rich user experiences through smartphones, AR and VR devices, and to provide robust simultaneous localization and mapping (SLAM) capabilities for robots and autonomous vehicles,” said Shin Yoon-seop, senior engineer at LG Electronics. “Our collaboration with CEVA has enabled us to address this market need by delivering a fully functional, compact 3D module.” “This compact 3D module will deliver exceptional performance thanks to our algorithms and the CEVA-XM4 imaging and vision DSP platform,” he said.

LG Electronics will present and demonstrate its smart 3D module at the CEVA Technology Symposium Series, taking place in China and Taiwan from October 23rd to 27th. Attendees will have the opportunity to interact with computer vision experts from CEVA and LG. More information and registration details can be found on the official symposium website.

CEVA's latest imaging and vision DSP platforms address the massive signal processing requirements of the most sophisticated machine learning and machine vision applications used in smartphones, surveillance cameras, augmented reality, drone sensing, and autonomous vehicles, all at low power. These DSP-based platforms feature a hybrid architecture comprised of scalar and vector DSP processors, along with a comprehensive Application Development Kit (ADK) to streamline software development.

The CEVA ADK includes: ▲CEVA-Link for seamless software-level integration with the host processor ▲a library of widely used optimized software algorithms ▲CEVA's second-generation deep neural network software framework (CEVA Deep Neural Network, CDNN2) that enables real-time machine learning with a fraction of the power consumption of modern GPU-based systems ▲state-of-the-art development and debugging tools. For more information, please visit the CEVA official website.