Stereo Camera Image Processing for Low-Resource Depth Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Monocular camera systems face high error rates in depth estimation for self-moving objects, while stereo camera systems consume excessive resources for image processing.
Innovation Solution
Process alternating images from a stereo camera using monocular image processing methods to derive information, reducing resource consumption and eliminating the need for simultaneous stereo image processing, while leveraging the fixed distance between cameras for improved depth estimation and object detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If stereo camera systems are used for depth estimation, then measurement precision is improved, but use of energy by stationary object increases
Solution Approach 1:
The patent segments the image processing task by alternating between left and right camera images across successive time points, processing only one camera's image at each time point rather than simultaneously processing both stereo images, thereby reducing computational load and energy consumption while maintaining depth estimation capability
Solution Approach 2:
The patent implements periodic action by alternating the processing of left and right camera images in a time-multiplexed manner, where at time point t1 the left camera image is processed, at time point t2 the right camera image is processed, creating a periodic pattern that reduces instantaneous resource requirements while maintaining stereo depth estimation
2Measurement precision
If stereo camera systems are used for depth estimation, then measurement precision is improved, but loss of substance increases
Solution Approach 1:
The patent segments the data processing stream by handling left and right camera images separately at different time points, processing one image at a time rather than simultaneously processing both stereo images, thereby reducing bandwidth requirements and memory consumption while maintaining depth estimation functionality
Solution Approach 2:
The patent applies periodic action by alternating the input of left and right camera images in a time-multiplexed sequence, where only one image is actively processed at each time point, reducing the instantaneous bandwidth and memory resources required compared to simultaneous stereo image processing
3Use of energy by stationary object
If monocular image processing is used, then use of energy by stationary object is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent merges the advantages of monocular processing (low energy consumption) with stereo processing (high accuracy) by alternating between left and right camera images in a time-multiplexed manner, effectively combining information from both cameras while maintaining the low resource consumption characteristic of monocular processing
Solution Approach 2:
The patent transitions from spatial parallel processing (simultaneous stereo image processing) to temporal sequential processing, adding a time dimension to the processing architecture, where left and right images are processed at different time points rather than simultaneously, thereby achieving stereo depth estimation with monocular-level resource efficiency
Data Source
AI summary
A method for deriving at least one item of information from images of a stereo camera. A resource-saving and at the same time accurate derivation of information from the images is achieved by alternately using and processing the images with methods of monocular image processing and deriving at least one item of information from the results. A motor vehicle comprising a stereo camera and comprising a control device which carries out the method, are also described.