3D Vision Processor Unit for Mobile Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional computing devices, especially mobile and embedded devices, face challenges in processing 3D vision algorithms in real-time due to high power consumption, limited processing capability, and memory constraints, making it impractical for applications like augmented reality and gesture recognition.
Innovation Solution
A 3D Vision Processor Unit (3VPU) is introduced, which pre-processes data between a camera and a System on a Chip (SoC) in mobile devices, utilizing silicon hardware to implement commonly used computational equations and complemented by software drivers, optimizing power consumption, flexibility, and memory usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional CPU/GPU is used for 3D vision processing, then processing capability is improved, but power consumption increases
Solution Approach 1:
The system segments processing tasks between the 3VPU (handling 3D vision processing) and the main CPU/GPU (handling other system tasks). This segmentation allows the mobile device to offload power-intensive 3D processing to a specialized unit, reducing overall power consumption while maintaining processing capability.
Solution Approach 2:
The 3VPU acts as an intermediary component between the camera and the main processor. It pre-processes 3D vision data before transmission to the CPU/GPU, reducing the computational burden on the main processor and enabling real-time processing with lower power consumption.
2Productivity
If high-end CPU processor is used, then processing capability is improved, but device size and cost increase
Solution Approach 1:
The processing system is segmented into a specialized 3VPU for 3D vision tasks and a standard ARM processor for general purposes. This allows the device to use a smaller, lower-power processor while achieving high-end 3D processing capability through the dedicated hardware unit.
Solution Approach 2:
The patent replaces the need for a large, high-end CPU with a specialized 3VPU hardware unit. This substitution enables 3D vision processing without requiring a bulky or expensive high-performance processor, thus reducing device size and cost.
3Speed
If real-time 3D processing is implemented, then processing speed is improved, but memory usage increases
Solution Approach 1:
The 3VPU extracts and processes only the essential 3D vision data in real-time, filtering out unnecessary information before transmission to the main system. This extraction approach enables real-time processing while minimizing the amount of data that needs to be stored in memory.
Solution Approach 2:
The 3VPU performs preliminary processing of 3D vision data, including filtering and feature extraction, before the data is transmitted to the main processor. This preliminary action reduces the volume of data that needs to be held in memory, enabling real-time processing with reduced memory requirements.
4Productivity
If custom hardware chip is used for 3D processing, then processing capability is improved, but device complexity and cost increase
Solution Approach 1:
The 3VPU is designed as a universal processing unit that can handle multiple 3D vision processing algorithms and functions. This multi-functionality reduces the need for separate custom hardware chips for different processing tasks, thereby reducing overall device complexity while maintaining high processing capability.
Data Source
AI summary
Methods and apparatuses are described for processing 3D vision algorithms. A 3D vision processor device comprises one or more 3D vision processing cores. Each 3D vision processing core includes one or more memory blocks for storing location values associated with 3D point cloud images and an arithmetic logic unit coupled to the one or more memory modules. The arithmetic logic unit includes a plurality of memory registers for temporarily storing location values associated with a point in a 3D point cloud image and a processing unit coupled to the plurality of memory registers for performing arithmetic operations on the location values stored in the memory registers, the arithmetic operations used for 3D vision processing algorithms. The 3D vision processing core also includes a communication link for transferring data between the arithmetic logic unit and the memory modules.