Optical flow estimation method and device based on pulse camera, computer equipment and medium
A pulse and optical flow technology, applied in the field of computer vision, can solve problems such as being easily affected by noise, rough optical flow estimation results, and difficulty in obtaining motion details, etc., achieving good robustness and improving accuracy
Inactive Publication Date: 2021-08-31
PEKING UNIV
View PDF2 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, conventional methods are often aimed at optical flow estimation of image sequences (optical flow estimation between two image frames), and the obtained optical flow estimation results are often rough, difficult to obtain motion details, and are also susceptible to noise, which is urgently needed. to improve or optimize
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0049] A method and device for estimating optical flow based on a pulse camera provided by the present invention will be explained and described in detail below in conjunction with the accompanying drawings.
[0050] Optical flow estimation has always been the focus of research in the field of computer vision. Substantial differences exist between the traditional image sequence-based optical flow estimation method, and the present invention innovatively provides an optical flow estimation method based on a pulse sequence collected by a pulse camera. It should be understood that the optical flow specifically refers to the amount of movement of pixels representing the same object (object) in one frame of the video image to the next frame, which is represented by a two-dimensional vector.
[0051]The pulse camera used in the present invention can specifically be a bionic pulse camera. In an impulse camera, each pixel independently and continuously receives light and accumulates ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention provides an optical flow estimation method and device based on a pulse camera, computer equipment and a medium. The optical flow estimation method comprises the following steps: acquiring pulse sequence collected by a pulse camera, and determining the pulse interval of the pulses based on the time when each pulse begins to accumulate and the time when the pulse is emitted; determining the pulse density corresponding to each time point in the time period to be subjected to optical flow estimation according to the pulse interval; and performing optical flow estimation by using the pulse density consistency constraint condition and the spatial domain local smoothness constraint condition to obtain an optical flow field. The optical flow field at any moment can be obtained from the pulse sequence, an optical flow field sequence is formed through the optical flow fields, and a coherent movement process is described. According to the invention, the influence of noise is reduced based on the illumination consistency along the motion trail among a plurality of pulse frame moments, and the optical flow field sequence with better robustness is estimated. The method is suitable for high-speed motion scenes, and is an important basis for application of image reconstruction, object tracking, object recognition and the like.
Description
technical field [0001] The present invention relates to the technical field of computer vision, and more specifically, the present invention can specifically provide a pulse camera-based optical flow estimation method and device, computer equipment, and media. Background technique [0002] Due to the presence of object motion, the information recorded by the sensor pixels at different times corresponds to different points on the object. How to analyze the motion structure of the recorded object more accurately and establish the motion relationship of image pixels at different time points is very important for image reconstruction, tracking and recognition of high-speed moving objects and other applications. Optical flow is a representation of the relative motion relationship of pixels, so how to estimate optical flow is of great research value. However, conventional methods are often aimed at optical flow estimation of image sequences (optical flow estimation between two im...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/207G01S11/12
CPCG06T7/207G01S11/12
Inventor 熊瑞勤赵菁黄铁军
Owner PEKING UNIV