| Title: |
SYSTEM AND METHOD FOR OBTAINING CAMERA PARAMETERS FROM MULTIPLE IMAGES AND COMPUTER PROGRAM PRODUCTS THEREOF |
| Document Number: |
20110128354 |
| Publication Date: |
June 2, 2011 |
| Appl. No: |
12/637369 |
| Application Filed: |
December 14, 2009 |
| Abstract: |
Systems and methods for obtaining camera parameters from images are provided. First, a sequence of original images associated with a target object under circular motion is obtained. Then, a background image and a foreground image corresponding to the target object within each original image are segmented. Next, shadow detection is performed for the target object within each original image. A first threshold and a second threshold are respectively determined according to the corresponding background and foreground images. Each original image, the corresponding background image, the first and second threshold are used for obtaining silhouette data and feature information associated with the target object within each original image. At least one camera parameter is obtained based on the entire feature information and the geometry of circular motion. |
| Inventors: |
TIEN, Tzu-Chieh (Zhongli City, TW); Huang, Po-Hao (Fengshan City, TW); Cheng, Chia-Ming (Hsinchu City, TW); Yang, Hao-Liang (Pingtung City, TW); Chen, Hsiao-Wei (Xinying City, TW); Lai, Shang-Hong (Hsinchu City, TW); Dong, Susan (Renwu Township, TW); Liu, Cheng-Da (Taipei City, TW); Tsai, Te-Lu (Taipei City, TW); Hsiao, Jung-Hsin (Xindian City, TW) |
| Claim: |
1. A system for obtaining camera parameters from a plurality of images, comprising: a processing module for obtaining a sequence of original images having a plurality of original images, segmenting a background image and a foreground image corresponding to a target object within each original image, performing shadow detection for the target object within each original image, determining a first threshold and a second threshold according to the corresponding background and foreground images, obtaining silhouette data by using each original image, the corresponding background image and the corresponding first threshold, and obtaining feature information associated with the target object within each original image by using each original image and the corresponding second threshold, wherein each original image within the sequence of original images is obtained by sequentially capturing the target object under circular motion and the silhouette data corresponds to the target object within each original image; and a calculation module for obtaining at least one camera parameter associated with the original images based on the entire feature information of the sequence of original images and the geometry of circular motion. |
| Claim: |
2. The system as claimed in claim 1, wherein the at least one camera parameter at least comprises an intrinsic parameter and/or an extrinsic parameter, the intrinsic parameter comprises at least one of a focal length, an aspect ratio, and a central location of each original image, and the extrinsic parameter is obtained according to the intrinsic parameter and the sequence of original images and is at least one of an image capture angle and an image capture position when capturing the target object. |
| Claim: |
3. The system as claimed in claim 1, further comprising: an image capturing unit for generating the sequence of original images by capturing the target object when the target object is under circular motion. |
| Claim: |
4. The system as claimed in claim 3, wherein the image capturing unit generates the sequence of original images by capturing the target object when the target object under the circular motion is at every constant angle. |
| Claim: |
5. The system as claimed in claim 1, further comprising: an integration module for constructing a three-dimensional model corresponding to the target object according to the silhouette data of the sequence of original images and the at least one camera parameter. |
| Claim: |
6. The system as claimed in claim 1, wherein the first threshold is obtained according to a shadow region of each original image and the corresponding background image, and the second threshold is obtained according to the shadow region of each original image and the corresponding foreground image. |
| Claim: |
7. The system as claimed in claim 1, wherein the processing module segments the background image and the foreground image corresponding to each original image by using a probability density function. |
| Claim: |
8. The system as claimed in claim 1, further comprising: an integration module for performing a calibration process on the original images according to the at least one camera parameter and constructing a three-dimensional model corresponding to the target object according to the calibrated original images and the at least one camera parameter. |
| Claim: |
9. The system as claimed in claim 1, wherein the processing module filters the background image by subtracting the first threshold from the background image and obtains the silhouette data according to each original image and the filtered background image. |
| Claim: |
10. The system as claimed in claim 1, wherein the processing module obtains the feature information associated with the target object within each original image by subtracting the second threshold from each original image. |
| Claim: |
11. A method for obtaining camera parameters from a plurality of images, comprising: obtaining a sequence of original images having a plurality of original images, wherein each original image within the sequence of original images is obtained by sequentially capturing a target object under circular motion; segmenting a background image and a foreground image corresponding to the target object within each original image; performing shadow detection for the target object within each original image and determining a first threshold and a second threshold according to the corresponding background and foreground images; obtaining silhouette data by using each original image, the corresponding background image and the corresponding first threshold, wherein the silhouette data corresponds to the target object within each original image; obtaining feature information associated with the target object within each original image by using each original image and the corresponding second threshold; and obtaining at least one camera parameter associated with the original images based on the entire feature information of the sequence of original images and the geometry of circular motion. |
| Claim: |
12. The method as claimed in claim 11, wherein the at least one camera parameter at least comprises an intrinsic parameter and/or an extrinsic parameter, the intrinsic parameter comprises at least one of a focal length, an aspect ratio, and a central location of each original image, and the extrinsic parameter is obtained according to the intrinsic parameter and the sequence of original images and is at least one of an image capture angle and an image capture position when capturing the target object. |
| Claim: |
13. The method as claimed in claim 11, further comprising: providing an image capturing unit for generating the sequence of original images by capturing the target object when the target object is under circular motion. |
| Claim: |
14. The method as claimed in claim 13, wherein the image capturing unit generates the sequence of original images by capturing the target object when the target object under the circular motion is at every constant angle. |
| Claim: |
15. The method as claimed in claim 11, further comprising: constructing a three-dimensional model corresponding to the target object according to the silhouette data of the sequence of original images and the at least one camera parameter. |
| Claim: |
16. The method as claimed in claim 11, wherein the first threshold is obtained according to a shadow region of each original image and the corresponding background image and the second threshold is obtained according to the shadow region of each original image and the corresponding foreground image. |
| Claim: |
17. The method as claimed in claim 11, further comprising: performing a calibration process on the original images according to the at least one camera parameter and constructing a three-dimensional model corresponding to the target object according to the calibrated original images and the at least one camera parameter. |
| Claim: |
18. The method as claimed in claim 11, wherein the background image is filtered by subtracting the first threshold from the background image and the silhouette data is obtained according to each original image and the filtered background image. |
| Claim: |
19. The method as claimed in claim 11, wherein the feature information associated with the target object within each original image is obtained by subtracting the second threshold from each original image. |
| Claim: |
20. A computer program product for being loaded by a machine to execute a method for obtaining camera parameters from a plurality of images, comprising: a first program code for obtaining a sequence of original images having a plurality of original images, wherein each original image within the sequence of original images is obtained by sequentially capturing a target object under circular motion via an image capturing unit; a second program code for segmenting a background image and a foreground image corresponding to the target object within each original image; a third program code for performing shadow detection for the target object within each original image and determining a first threshold and a second threshold according to the corresponding background and foreground images; a fourth program code for obtaining silhouette data by using each original image, the corresponding background image and the corresponding first threshold, wherein the silhouette data corresponds to the target object within each original image; a fifth program code for obtaining feature information associated with the target object within each original image by using each original image and the corresponding second threshold; and a sixth program code for obtaining at least one camera parameter associated with the original images based on the entire feature information of the sequence of original images and the geometry of circular motion. |
| Current U.S. Class: |
348/50 |
| Current International Class: |
04; 06; 06 |
| Accession Number: |
edspap.20110128354 |
| Database: |
USPTO Patent Applications |
