期刊
COMPUTER VISION - ECCV 2022, PT I
卷 13661, 期 -, 页码 38-54出版社
SPRINGER INTERNATIONAL PUBLISHING AG
DOI: 10.1007/978-3-031-19769-7_3
关键词
Crowd localization; Crowd counting; Transformer
资金
- National Key R&D Program of China [2018YFB1004602]
In this paper, an elegant and end-to-end Crowd Localization Transformer (CLTR) is proposed to solve the task of crowd localization. The proposed method treats crowd localization as a direct set prediction problem and introduces a KMO-based Hungarian matcher to reduce ambiguous points and generate more reasonable matching results. Extensive experiments demonstrate the effectiveness of the proposed method.
Crowd localization, predicting head positions, is a more practical and high-level task than simply counting. Existing methods employ pseudo-bounding boxes or pre-designed localization maps, relying on complex post-processing to obtain the head positions. In this paper, we propose an elegant, end-to-end Crowd Localization TRansformer named CLTR that solves the task in the regression-based paradigm. The proposed method views the crowd localization as a direct set prediction problem, taking extracted features and trainable embeddings as input of the transformer-decoder. To reduce the ambiguous points and generate more reasonable matching results, we introduce a KMO-based Hungarian matcher, which adopts the nearby context as the auxiliary matching cost. Extensive experiments conducted on five datasets in various data settings show the effectiveness of our method. In particular, the proposed method achieves the best localization performance on the NWPU-Crowd, UCF-QNRF, and ShanghaiTech Part A datasets.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据