Depth Sensor Pixel With Four Taps For Dynamic Range And Speed
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current depth sensors face a trade-off between dynamic range and operating speed, requiring multiple exposures and read operations to generate accurate depth maps, which limits their performance.
Innovation Solution
A depth sensor design that includes a pixel structure with multiple taps and an overflow transistor, allowing charges to be stored during different integration periods and read out through a single operation, enabling improved dynamic range and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple exposures and multiple read operations are performed to obtain charge information at different exposure times, then the dynamic range of the depth map is improved, but the operating speed of the depth sensor decreases
Solution Approach 1:
The pixel is divided into four separate taps (first, second, third, and fourth taps), each capable of independently storing charge information from different integration periods. This segmentation allows simultaneous capture of multiple exposure times without requiring sequential read operations, thereby improving both dynamic range and operating speed.
Solution Approach 2:
Multiple charge storage functions are merged into a single pixel structure with four taps. All four taps can store charge information from different integration periods simultaneously, and a single read operation can retrieve information from all taps, eliminating the need for multiple sequential read operations.
2Productivity
If a single read operation is used to improve operating speed, then the operating speed of the depth sensor is improved, but the dynamic range of the depth map decreases
Solution Approach 1:
The pixel is divided into four separate taps (first, second, third, and fourth taps), each capable of independently storing charge information from different integration periods. This segmentation allows simultaneous capture of multiple exposure times without requiring sequential read operations, thereby improving both dynamic range and operating speed.
Solution Approach 2:
Each tap within the pixel is designed to be multi-functional, capable of storing charge information from any integration period. This universality allows a single read operation to access diverse exposure time information, maintaining high dynamic range while achieving fast operating speed.
3Measurement precision
If multiple read operations are performed to capture charge information at different exposure times, then the dynamic range is improved, but the complexity of the readout process increases
Solution Approach 1:
Multiple charge storage functions are merged into a single pixel structure with four taps. All four taps can store charge information from different integration periods simultaneously, and a single read operation can retrieve information from all taps, eliminating the need for multiple sequential read operations.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the dynamic range of depth maps and increases the operating speed of the depth sensor by allowing both shorter and longer exposure information to be captured in a single read operation, improving the sensing of both short-range and long-range objects.
Implementation Method 1
a photoelectric conversion device connected in common with the first tap, the second tap, the third tap, and the fourth tap
Data Source
AI summary
Provided is a depth sensor which includes a pixel and a row driver that controls the pixel, the pixel including a first tap, a second tap, a third tap, and a fourth tap, an overflow transistor, and a photoelectric conversion device. Each of the first tap, the second tap, the third tap, and the fourth tap includes a photo transistor, a transfer transistor, and a readout circuit. In a first integration period of a global mode, the row driver activates a second photo gate signal controlling the photo transistor of the second tap and a third photo gate signal controlling the photo transistor of the third tap. In a second integration period of the global mode, the row driver activates a first photo gate signal controlling the photo transistor of the first tap and a fourth photo gate signal controlling the photo transistor of the fourth tap.


