A real-time photometric stereo vision method and device based on an event camera

By designing the illumination patterns of the event camera and using a null vector fusion method, combined with SVD and deep learning, real-time reconstruction of photometric stereo vision using the event camera was achieved. This solved the problems of complexity, time consumption, and insufficient information prediction in traditional methods, and improved the quality of normal estimation and the speed of data acquisition.

CN118781255BActive Publication Date: 2026-01-06PEKING UNIV
2 Cites 0 Cited by

Patent Information

Application Number
CN202410769824.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-14
Publication Date
2026-01-06
Estimated Expiration
2044-06-14

AI Technical Summary

Technical Problem

Traditional photometric stereo vision methods based on frame cameras are complex and time-consuming, making it difficult to meet real-time requirements. Hybrid camera systems have insufficient prediction of information in overexposed areas and cannot predict information in underexposed areas, resulting in limited dynamic range reconstruction by existing methods.

Method used

By leveraging the unique properties of event cameras, a lighting pattern is designed. Lighting signals are synchronously acquired through the event camera, null vectors are constructed, and surface normals are solved by fusing them. Surface normal estimation is optimized by combining SVD and deep learning methods, and real-time reconstruction is performed using an event camera photometric stereo vision neural network.

Benefits of technology

It enables real-time photometric stereo vision using a single event camera, reducing the cost of multiple cameras and synchronization, improving data sampling density and normal estimation quality, and accurately acquiring information about high-contrast and high-specular-reflection objects.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

A real-time photometric stereo vision method and device based on an event camera, belonging to the field of computer vision, includes: lighting pattern design; simultaneously acquiring lighting signals and synchronization signals generated by lighting changes in the test object using an event camera; using the synchronization signals to calibrate the lighting direction when other events occur, achieving synchronous acquisition; constructing null vectors directly related to the surface normals of the test object from each pair of consecutively occurring events; and fusing the null vectors to solve for the surface normals. This invention reconstructs the surface normals of the test object from the basic model triggered by the event camera, reducing the cost of multiple cameras and synchronization; it utilizes the high event resolution of the event camera to acquire data under continuously changing light sources, increasing data sampling density and improving the quality of normal estimation; and it leverages the high dynamic range of the event camera to improve data acquisition speed and quality through native high dynamic range and compressed event representation.
Need to check novelty before this filing date? Find Prior Art