The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Kadar bingkai tinggi dan kelewatan ultra-rendah adalah keperluan paling penting untuk membina sistem interaksi manusia-mesin yang sangat baik. Sebagai algoritma pengesanan titik kunci tempatan dan pengekstrakan ciri terkini, A-KAZE menunjukkan ketepatan dan keteguhan yang tinggi. Ruang skala bukan linear ialah salah satu modul terpenting dalam A-KAZE, tetapi ia bukan sahaja mempunyai sekurang-kurangnya satu kelewatan bingkai dan juga tidak mesra perkakasan. Kertas ini mencadangkan ruang skala tak linear berorientasikan perkakasan untuk kadar bingkai tinggi dan sistem pemadanan A-KAZE kelewatan ultra-rendah. Dalam sistem padanan yang dicadangkan, satu bahagian ruang skala tak linear dihadapkan secara sementara dan dikira dalam bingkai sebelumnya (cadangan #1), supaya kelewatan pemprosesan dikurangkan kepada kurang daripada 1 ms. Untuk meningkatkan ketepatan padanan yang dipengaruhi oleh cadangan #1, prapelarasan skala tak linear (cadangan #2) dicadangkan. Dua bingkai sebelumnya digunakan untuk melakukan anggaran gerakan untuk meramalkan vektor gerakan antara bingkai sebelumnya dan bingkai semasa. Untuk meningkatkan lagi ketepatan padanan, prapelarasan tahap piksel (cadangan #3) dicadangkan. Prapelarasan berubah daripada peringkat blok kepada peringkat piksel, setiap piksel diberikan vektor gerakan yang unik. Keputusan eksperimen membuktikan bahawa sistem padanan yang dicadangkan menunjukkan ketepatan padanan purata lebih tinggi daripada 95% iaitu 5.88% lebih tinggi daripada kadar bingkai tinggi sedia ada dan sistem padanan kelewatan ultra-rendah. Bagi prestasi perkakasan, sistem padanan yang dicadangkan memproses video VGA (640×480 piksel/bingkai) pada kelajuan 784 bingkai/saat (fps) dengan kelewatan 0.978 ms/bingkai.
Songlin DU
Southeast University,Ministry of Education
Yuan LI
Waseda University
Takeshi IKENAGA
Waseda University
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.
Salinan
Songlin DU, Yuan LI, Takeshi IKENAGA, "Temporally Forward Nonlinear Scale Space for High Frame Rate and Ultra-Low Delay A-KAZE Matching System" in IEICE TRANSACTIONS on Information,
vol. E103-D, no. 6, pp. 1226-1235, June 2020, doi: 10.1587/transinf.2019MVP0019.
Abstract: High frame rate and ultra-low delay are the most essential requirements for building excellent human-machine-interaction systems. As a state-of-the-art local keypoint detection and feature extraction algorithm, A-KAZE shows high accuracy and robustness. Nonlinear scale space is one of the most important modules in A-KAZE, but it not only has at least one frame delay and but also is not hardware friendly. This paper proposes a hardware oriented nonlinear scale space for high frame rate and ultra-low delay A-KAZE matching system. In the proposed matching system, one part of nonlinear scale space is temporally forward and calculated in the previous frame (proposal #1), so that the processing delay is reduced to be less than 1 ms. To improve the matching accuracy affected by proposal #1, pre-adjustment of nonlinear scale (proposal #2) is proposed. Previous two frames are used to do motion estimation to predict the motion vector between previous frame and current frame. For further improvement of matching accuracy, pixel-level pre-adjustment (proposal #3) is proposed. The pre-adjustment changes from block-level to pixel-level, each pixel is assigned an unique motion vector. Experimental results prove that the proposed matching system shows average matching accuracy higher than 95% which is 5.88% higher than the existing high frame rate and ultra-low delay matching system. As for hardware performance, the proposed matching system processes VGA videos (640×480 pixels/frame) at the speed of 784 frame/second (fps) with a delay of 0.978 ms/frame.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2019MVP0019/_p
Salinan
@ARTICLE{e103-d_6_1226,
author={Songlin DU, Yuan LI, Takeshi IKENAGA, },
journal={IEICE TRANSACTIONS on Information},
title={Temporally Forward Nonlinear Scale Space for High Frame Rate and Ultra-Low Delay A-KAZE Matching System},
year={2020},
volume={E103-D},
number={6},
pages={1226-1235},
abstract={High frame rate and ultra-low delay are the most essential requirements for building excellent human-machine-interaction systems. As a state-of-the-art local keypoint detection and feature extraction algorithm, A-KAZE shows high accuracy and robustness. Nonlinear scale space is one of the most important modules in A-KAZE, but it not only has at least one frame delay and but also is not hardware friendly. This paper proposes a hardware oriented nonlinear scale space for high frame rate and ultra-low delay A-KAZE matching system. In the proposed matching system, one part of nonlinear scale space is temporally forward and calculated in the previous frame (proposal #1), so that the processing delay is reduced to be less than 1 ms. To improve the matching accuracy affected by proposal #1, pre-adjustment of nonlinear scale (proposal #2) is proposed. Previous two frames are used to do motion estimation to predict the motion vector between previous frame and current frame. For further improvement of matching accuracy, pixel-level pre-adjustment (proposal #3) is proposed. The pre-adjustment changes from block-level to pixel-level, each pixel is assigned an unique motion vector. Experimental results prove that the proposed matching system shows average matching accuracy higher than 95% which is 5.88% higher than the existing high frame rate and ultra-low delay matching system. As for hardware performance, the proposed matching system processes VGA videos (640×480 pixels/frame) at the speed of 784 frame/second (fps) with a delay of 0.978 ms/frame.},
keywords={},
doi={10.1587/transinf.2019MVP0019},
ISSN={1745-1361},
month={June},}
Salinan
TY - JOUR
TI - Temporally Forward Nonlinear Scale Space for High Frame Rate and Ultra-Low Delay A-KAZE Matching System
T2 - IEICE TRANSACTIONS on Information
SP - 1226
EP - 1235
AU - Songlin DU
AU - Yuan LI
AU - Takeshi IKENAGA
PY - 2020
DO - 10.1587/transinf.2019MVP0019
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 6
JA - IEICE TRANSACTIONS on Information
Y1 - June 2020
AB - High frame rate and ultra-low delay are the most essential requirements for building excellent human-machine-interaction systems. As a state-of-the-art local keypoint detection and feature extraction algorithm, A-KAZE shows high accuracy and robustness. Nonlinear scale space is one of the most important modules in A-KAZE, but it not only has at least one frame delay and but also is not hardware friendly. This paper proposes a hardware oriented nonlinear scale space for high frame rate and ultra-low delay A-KAZE matching system. In the proposed matching system, one part of nonlinear scale space is temporally forward and calculated in the previous frame (proposal #1), so that the processing delay is reduced to be less than 1 ms. To improve the matching accuracy affected by proposal #1, pre-adjustment of nonlinear scale (proposal #2) is proposed. Previous two frames are used to do motion estimation to predict the motion vector between previous frame and current frame. For further improvement of matching accuracy, pixel-level pre-adjustment (proposal #3) is proposed. The pre-adjustment changes from block-level to pixel-level, each pixel is assigned an unique motion vector. Experimental results prove that the proposed matching system shows average matching accuracy higher than 95% which is 5.88% higher than the existing high frame rate and ultra-low delay matching system. As for hardware performance, the proposed matching system processes VGA videos (640×480 pixels/frame) at the speed of 784 frame/second (fps) with a delay of 0.978 ms/frame.
ER -