3D sketching for 3D object retrieval
- PDF / 2,399,473 Bytes
- 27 Pages / 439.642 x 666.49 pts Page_size
- 46 Downloads / 267 Views
3D sketching for 3D object retrieval Bo Li1
· Juefei Yuan1 · Yuxiang Ye2 · Yijuan Lu2 · Chaoyang Zhang3 · Qi Tian4
Received: 1 May 2020 / Revised: 30 August 2020 / Accepted: 6 October 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract Sketching provides the most natural way to provide a visual search query for visual object search. However, how to draw 3D sketches in a three-dimensional space and how to use a hand-drawn 3D sketch to search similar 3D models are not only interesting and novel, but also challenging research topics. In this paper, we try to answer them by initiating a novel study on 3D sketching and build a 3D sketching system which allows users to freely draw 3D sketches in the air and demonstrate its promising potentials in related applications such as collecting 3D sketch data and conducting 3D sketch-based 3D model retrieval. By utilizing the 3D sketching system, we collect a 3D sketch dataset, build a 3D sketch-based 3D model retrieval benchmark, and organize a Eurographics Shape Retrieval Contest (SHREC) track on 3D sketch-based shape retrieval based on the benchmark. We investigate 3D sketch and model matching problems and propose a novel 3D sketch-based model retrieval algorithm CNN-SBR based on Convolutional Neural Networks (CNNs) and achieve the best performance in the SHREC track. We wish that the 3D sketching system, the 3D sketch-based model retrieval benchmark, and the proposed 3D sketch-based model retrieval algorithm CNN-SBR will further promote sketch-based shape retrieval and its applications. We have made all of these publicly available on the project homepage: http://orca.st.usm.edu/∼bli/ SBR16/project.html. Keywords 3D sketching · Kinect · Sketch-based 3D model retrieval · Convolutional neural networks
1 Introduction Content-based 3D shape retrieval [66] is important for many various related applications such as computer-aided design (CAD), 3D movie and game production, augmented reality (AR), virtual reality (VR), and 3D printin g. Given a query which is often a 2D sketch/image or a 3D model, content-based 3D model retrieval is to retrieve relevant 3D models (typically
Yijuan Lu
[email protected]; [email protected]
Extended author information available on the last page of the article.
Multimedia Tools and Applications
only single object models) coming from the same category as the query based on a similarity/distance metric. Effectiveness, efficiency, and scalability are the three most important performance aspects, which can be measured by a set of retrieval performance evaluation metrics [39, 59] that are commonly used in the field of information retrieval. Due to the intuitiveness, convenience and potential for related applications, sketch-based 3D shape retrieval has received a lot of attentions from both inside and outside of the community of 3D shape retrieval. As we know, as a universal form of communication to depict the visual world, sketching has been used by human beings since tens of thousands of years ago [22]. Nowadays, s
Data Loading...
