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
Dalam kertas kerja ini, kami menambah baik ketepatan kaedah penjanaan tracklet sedia ada dengan membaiki tracklet berdasarkan penilaian kualiti dan penyebaran pengesanannya. Bermula daripada pengesanan objek, kami menjana tracklet menggunakan tiga kaedah sedia ada. Kemudian kami melakukan penilaian kualiti tracklet bersama untuk menjaringkan setiap tracklet dan menapis tracklet yang baik berdasarkan markah mereka. Kaedah penyebaran pengesanan direka untuk memindahkan pengesanan dalam tracklet yang baik kepada yang buruk untuk membaiki tracklet yang buruk. Penilaian kualiti tracklet dalam kaedah kami dilaksanakan oleh konsistensi pengesanan intra-tracklet dan kesempurnaan pengesanan antara tracklet. Dua kaedah pembiakan; penyebaran global dan penyebaran tempatan ditakrifkan untuk mencapai penyebaran tracklet yang lebih tepat. Kami menunjukkan keberkesanan kaedah yang dicadangkan pada dataset MOT 15
Nii L. SOWAH
University of Electronic Science and Technology of China
Qingbo WU
University of Electronic Science and Technology of China
Fanman MENG
University of Electronic Science and Technology of China
Liangzhi TANG
University of Electronic Science and Technology of China
Yinan LIU
University of Electronic Science and Technology of China
Linfeng XU
University of Electronic Science and Technology of China
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
Nii L. SOWAH, Qingbo WU, Fanman MENG, Liangzhi TANG, Yinan LIU, Linfeng XU, "A Propagation Method for Multi Object Tracklet Repair" in IEICE TRANSACTIONS on Information,
vol. E101-D, no. 9, pp. 2413-2416, September 2018, doi: 10.1587/transinf.2018EDL8029.
Abstract: In this paper, we improve upon the accuracy of existing tracklet generation methods by repairing tracklets based on their quality evaluation and detection propagation. Starting from object detections, we generate tracklets using three existing methods. Then we perform co-tracklet quality evaluation to score each tracklet and filtered out good tracklet based on their scores. A detection propagation method is designed to transfer the detections in the good tracklets to the bad ones so as to repair bad tracklets. The tracklet quality evaluation in our method is implemented by intra-tracklet detection consistency and inter-tracklet detection completeness. Two propagation methods; global propagation and local propagation are defined to achieve more accurate tracklet propagation. We demonstrate the effectiveness of the proposed method on the MOT 15 dataset
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2018EDL8029/_p
Salinan
@ARTICLE{e101-d_9_2413,
author={Nii L. SOWAH, Qingbo WU, Fanman MENG, Liangzhi TANG, Yinan LIU, Linfeng XU, },
journal={IEICE TRANSACTIONS on Information},
title={A Propagation Method for Multi Object Tracklet Repair},
year={2018},
volume={E101-D},
number={9},
pages={2413-2416},
abstract={In this paper, we improve upon the accuracy of existing tracklet generation methods by repairing tracklets based on their quality evaluation and detection propagation. Starting from object detections, we generate tracklets using three existing methods. Then we perform co-tracklet quality evaluation to score each tracklet and filtered out good tracklet based on their scores. A detection propagation method is designed to transfer the detections in the good tracklets to the bad ones so as to repair bad tracklets. The tracklet quality evaluation in our method is implemented by intra-tracklet detection consistency and inter-tracklet detection completeness. Two propagation methods; global propagation and local propagation are defined to achieve more accurate tracklet propagation. We demonstrate the effectiveness of the proposed method on the MOT 15 dataset},
keywords={},
doi={10.1587/transinf.2018EDL8029},
ISSN={1745-1361},
month={September},}
Salinan
TY - JOUR
TI - A Propagation Method for Multi Object Tracklet Repair
T2 - IEICE TRANSACTIONS on Information
SP - 2413
EP - 2416
AU - Nii L. SOWAH
AU - Qingbo WU
AU - Fanman MENG
AU - Liangzhi TANG
AU - Yinan LIU
AU - Linfeng XU
PY - 2018
DO - 10.1587/transinf.2018EDL8029
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E101-D
IS - 9
JA - IEICE TRANSACTIONS on Information
Y1 - September 2018
AB - In this paper, we improve upon the accuracy of existing tracklet generation methods by repairing tracklets based on their quality evaluation and detection propagation. Starting from object detections, we generate tracklets using three existing methods. Then we perform co-tracklet quality evaluation to score each tracklet and filtered out good tracklet based on their scores. A detection propagation method is designed to transfer the detections in the good tracklets to the bad ones so as to repair bad tracklets. The tracklet quality evaluation in our method is implemented by intra-tracklet detection consistency and inter-tracklet detection completeness. Two propagation methods; global propagation and local propagation are defined to achieve more accurate tracklet propagation. We demonstrate the effectiveness of the proposed method on the MOT 15 dataset
ER -