Method and device for tracking dense depth images

A depth image and depth tracking technology, which is applied in the field of dense depth image tracking, can solve the problems of time-consuming calculation of optical flow field, complex tracking algorithm, long tracking time, etc.

Inactive Publication Date: 2010-09-29
TSINGHUA UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the existing algorithms, the calculation of the optical flow field is time-consuming, how to reduce its time complexity is a difficulty in current research
[0006] In the process of realizing the present invention, the inventor found that the prior art has at least the following problems: the existing dense depth image tracking technology has complex tracking algorithms and relatively long tracking time

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for tracking dense depth images
  • Method and device for tracking dense depth images
  • Method and device for tracking dense depth images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0094] see figure 1 , the present embodiment provides a method for dense depth image tracking, the method comprising:

[0095] 101: Obtain the p-th frame image and the q-th frame image of two adjacent frames of images, and the depth image of the p-th frame image is known;

[0096] Specifically, the frame number of the p-th frame image is p, and the frame number of the q-th frame image is q; when pq, the qth frame image is the previous frame image of the pth frame image, that is, backward tracking.

[0097] 102: For any pixel point (i, j) of the qth frame image, according to the color difference between the pixel point (i, j) and the pixel point (m, n) of the pth frame image and preset The standard deviation of the two-dimensional normal distribution is calculated to obtain the weight of the pixel point (m, n), the value range of (m, n) is centered on (i, j), and the default tracking value is the side length within the square tracking window of

[0098] Wherein, when calcula...

Embodiment 2

[0120] see figure 2 , the embodiment of the present invention provides another dense depth image tracking method, including:

[0121] 201: Obtain an image sequence and a depth image corresponding to a key frame image in the image sequence;

[0122] Wherein, the depth image refers to a grayscale image obtained by quantizing the distance of each pixel in the plane image from the camera in the real world to 0-255. Generally speaking, the pixel farthest from the camera in the image is quantized to 0, and the pixel closest to the camera is quantized to 255, and the distances of the remaining pixels have different quantization algorithms, such as: linear quantization, piecewise linear quantization, binary Subfunction quantization, etc.

[0123] Wherein, the image sequence includes key frame images and non-key frame images. A key frame image, in this embodiment of the present invention, refers to an image that already has a depth image before tracking starts, and may be any frame i...

example 1

[0148] Example 1: The input image sequence is the 1st to 100th frame of the head rotation, the key frame is set to the 1st frame, and the standard deviation parameter of the two-dimensional normal distribution is set to sigma=8, and the tracking window size parameter box_size=7. Since there is only one key frame and it is the first frame of the image sequence, the method of forward tracking is adopted. Get the first frame and its depth image and the second frame image, input the preset depth tracking algorithm, and calculate the depth image of the second frame; then take the second frame and its depth image and the third frame image, input the preset Depth tracking algorithm, calculate the depth image of the third frame; loop in turn until all the depth images of 100 frames are obtained.

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

No PUM Login to view more

Abstract

The invention discloses a method and a device for tracking dense depth images, and belongs to the field of stereo videos. The method comprises the following steps: acquiring two adjacent frame images, namely a pth frame image of which depth image is known and a qth frame image; for any pixel point (i, j) of the qth frame image, according to the color difference between the pixel point (i, j) and a pixel point (m, n) of the pth frame image and the preset two-dimensional normal distribution standard difference, acquiring the weight of the pixel point (m, n) by calculating, wherein the value range of the (m, n) uses the (i, j) as a center, and a preset tracking value is in a tracking window of a square side; and according to the weight of the pixel point (m, n), weighing and averaging the depths of all pixel points (m, n) in the tracking window to acquire the depth of the pixel point (i, j). The device comprises an image acquisition module, a weight calculating module and a depth tracking module. The method and the device have the effects of low calculation complexity and short tracking time.

Description

technical field [0001] The invention relates to the field of stereoscopic video, in particular to a method and device for dense depth image tracking. Background technique [0002] With the rapid development of computer technology, flat video can no longer meet people's needs, and stereoscopic video technology has developed by leaps and bounds. [0003] Stereoscopic video is a new video technology developed by studying the imaging principle of human visual system. Stereoscopic video is to study how to express the depth of the object in the video, that is, the distance between the object and the person in the real world, and obtain the three-dimensional stereoscopic video through the synthesis of the depth and the ordinary plane video. Therefore, an important element in the stereoscopic video production process is: the depth image. There are many ways to obtain a depth image, among which, to obtain a depth image based on a single-channel video, the depth image is generally o...

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
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): H04N13/00G06T7/00
Inventor 谢旭东黎政刘晓冬戴琼海
Owner TSINGHUA UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap