SHREC'22 track: Sketch-based 3D shape retrieval in the wild. (October 2022)

• Record Type: Journal Article
• Authors: Qin, Jie; Yuan, Shuaihang; Chen, Jiaxin; Ben Amor, Boulbaba; Fang, Yi; Hoang-Xuan, Nhat; Chu, Chi-Bien; Nguyen-Ngoc, Khoi-Nguyen; Cao, Thien-Tri; Ngo, Nhat-Khang; Huynh, Tuan-Luc; Nguyen, Hai-Dang; Tran, Minh-Triet; Luo, Haoyang; Wang, Jianning; Zhang, Zheng; Xin, Zihao; Wang, Yang; Wang, Feng; Tang, Ying; Chen, Haiqin; Wang, Yan; Zhou, Qunying; Zhang, Ji; Wang, Hongyuan
• Is Part Of: Computers & graphics. Volume 107(2022)
• Page Start: 104
• Page End: 115
• Publication Date: 2022-10
• Subjects: Sketch-based 3D shape retrieval -- Cross-modality retrieval -- Shape retrieval in the wild -- Point cloud classification; Computer graphics -- Periodicals; 006.6
• DOI: 10.1016/j.cag.2022.07.009
• Languages: English
• ISSNs: 0097-8493

Abstract

Abstract: Sketch-based 3D shape retrieval (SBSR) is an important yet challenging task, which has drawn more and more attention in recent years. Existing approaches address the problem in a restricted setting, without appropriately simulating real application scenarios. To mimic the realistic setting, in this track, we adopt large-scale sketches drawn by amateurs of different levels of drawing skills, as well as a variety of 3D shapes including not only CAD models but also models scanned from real objects. We define two SBSR tasks and construct two benchmarks consisting of more than 46, 000 CAD models, 1700 realistic models, and 145, 000 sketches in total. Four teams participated in this track and submitted 15 runs for the two tasks, evaluated by 7 commonly-adopted metrics. We hope that, the benchmarks, the comparative results, and the open-sourced evaluation code will foster future research in this direction among the 3D object retrieval community. Graphical abstract: Highlights: We propose a more challenging setting for SBSR in realistic application scenarios. We provide 2 benchmarks of 46k CAD models, 1.7k realistic models, and 145k sketches. Four teams participated in this track and submitted 15 runs for the two tasks. We hope the benchmark will foster future research in this direction in the community.