SketchANIMAR：基于草图的 3D 动物细粒度检索
计算机视觉与模式识别
2023-08-10 v2
摘要
近年来，3D 物体检索因其在计算机视觉、计算机图形学、虚拟现实和增强现实中的广泛应用而变得日益重要。然而，由于 3D 模型在形状、尺寸和纹理上各异，且包含大量多边形与顶点，3D 物体检索面临显著挑战。为此，我们引入了一项新颖的 SHREC 挑战赛道，专注于使用草图查询从数据集中检索相关的 3D 动物模型，并通过现有草图加速访问 3D 模型。此外，本研究构建了一个名为 ANIMAR 的新数据集，包含 711 个独特的 3D 动物模型与 140 条相应的草图查询。我们的竞赛要求参与者基于复杂而细致的草图检索 3D 模型。我们收到了来自八支队伍、共 204 次运行结果的满意表现。尽管仍需进一步改进，所提出的任务有潜力激励 3D 物体检索领域的更多研究，并可能为广泛应用带来益处。我们也对特征提取与匹配技术的改进以及创建更多样化数据集以评估检索性能等未来研究的潜在方向提供了见解。https://aichallenge.hcmus.edu.vn/sketchanimar
引用
@article{arxiv.2304.05731,
title = {SketchANIMAR: Sketch-based 3D Animal Fine-Grained Retrieval},
author = {Trung-Nghia Le and Tam V. Nguyen and Minh-Quan Le and Trong-Thuan Nguyen and Viet-Tham Huynh and Trong-Le Do and Khanh-Duy Le and Mai-Khiem Tran and Nhat Hoang-Xuan and Thang-Long Nguyen-Ho and Vinh-Tiep Nguyen and Nhat-Quynh Le-Pham and Huu-Phuc Pham and Trong-Vu Hoang and Quang-Binh Nguyen and Trong-Hieu Nguyen-Mau and Tuan-Luc Huynh and Thanh-Danh Le and Ngoc-Linh Nguyen-Ha and Tuong-Vy Truong-Thuy and Truong Hoai Phong and Tuong-Nghiem Diep and Khanh-Duy Ho and Xuan-Hieu Nguyen and Thien-Phuc Tran and Tuan-Anh Yang and Kim-Phat Tran and Nhu-Vinh Hoang and Minh-Quang Nguyen and Hoai-Danh Vo and Minh-Hoa Doan and Hai-Dang Nguyen and Akihiro Sugimoto and Minh-Triet Tran},
journal= {arXiv preprint arXiv:2304.05731},
year = {2023}
}
备注
Accepted to Computers & Graphics (3DOR 2023, Journal track)