This paper proposes a high-precision, high-speed keypoint matching method using two-stage randomized trees (RTs). The keypoint classification uses conventional RTs for high-precision, real-time keypoint matching. However, the wide variety of view transformations for templates expressed by RTs make it diffidult to achieve high-precision classification for all transformations with a single RTs. To solve this problem, the proposed method classifies the template view transformations in the first stage and then, in the second stage, classifies the keypoints using the RTs that corresponds to each of the view transformations classified in the first stage. Testing demonstrated that the proposed method is 88.5% more precise than SIFT, and 63.5% more precise than using conventional RTs for images in which the viewpoint of the object is rotated by 70 degrees. We have also shown that the proposed method supports real-time keypoint matching at 12 fps.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Shoichi SHIMIZU, Hironobu FUJIYOSHI, "Keypoint Recognition with Two-Stage Randomized Trees" in IEICE TRANSACTIONS on Information,
vol. E95-D, no. 7, pp. 1766-1774, July 2012, doi: 10.1587/transinf.E95.D.1766.
Abstract: This paper proposes a high-precision, high-speed keypoint matching method using two-stage randomized trees (RTs). The keypoint classification uses conventional RTs for high-precision, real-time keypoint matching. However, the wide variety of view transformations for templates expressed by RTs make it diffidult to achieve high-precision classification for all transformations with a single RTs. To solve this problem, the proposed method classifies the template view transformations in the first stage and then, in the second stage, classifies the keypoints using the RTs that corresponds to each of the view transformations classified in the first stage. Testing demonstrated that the proposed method is 88.5% more precise than SIFT, and 63.5% more precise than using conventional RTs for images in which the viewpoint of the object is rotated by 70 degrees. We have also shown that the proposed method supports real-time keypoint matching at 12 fps.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.E95.D.1766/_p
Copy
@ARTICLE{e95-d_7_1766,
author={Shoichi SHIMIZU, Hironobu FUJIYOSHI, },
journal={IEICE TRANSACTIONS on Information},
title={Keypoint Recognition with Two-Stage Randomized Trees},
year={2012},
volume={E95-D},
number={7},
pages={1766-1774},
abstract={This paper proposes a high-precision, high-speed keypoint matching method using two-stage randomized trees (RTs). The keypoint classification uses conventional RTs for high-precision, real-time keypoint matching. However, the wide variety of view transformations for templates expressed by RTs make it diffidult to achieve high-precision classification for all transformations with a single RTs. To solve this problem, the proposed method classifies the template view transformations in the first stage and then, in the second stage, classifies the keypoints using the RTs that corresponds to each of the view transformations classified in the first stage. Testing demonstrated that the proposed method is 88.5% more precise than SIFT, and 63.5% more precise than using conventional RTs for images in which the viewpoint of the object is rotated by 70 degrees. We have also shown that the proposed method supports real-time keypoint matching at 12 fps.},
keywords={},
doi={10.1587/transinf.E95.D.1766},
ISSN={1745-1361},
month={July},}
Copy
TY - JOUR
TI - Keypoint Recognition with Two-Stage Randomized Trees
T2 - IEICE TRANSACTIONS on Information
SP - 1766
EP - 1774
AU - Shoichi SHIMIZU
AU - Hironobu FUJIYOSHI
PY - 2012
DO - 10.1587/transinf.E95.D.1766
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E95-D
IS - 7
JA - IEICE TRANSACTIONS on Information
Y1 - July 2012
AB - This paper proposes a high-precision, high-speed keypoint matching method using two-stage randomized trees (RTs). The keypoint classification uses conventional RTs for high-precision, real-time keypoint matching. However, the wide variety of view transformations for templates expressed by RTs make it diffidult to achieve high-precision classification for all transformations with a single RTs. To solve this problem, the proposed method classifies the template view transformations in the first stage and then, in the second stage, classifies the keypoints using the RTs that corresponds to each of the view transformations classified in the first stage. Testing demonstrated that the proposed method is 88.5% more precise than SIFT, and 63.5% more precise than using conventional RTs for images in which the viewpoint of the object is rotated by 70 degrees. We have also shown that the proposed method supports real-time keypoint matching at 12 fps.
ER -