• Korean Journal of Computational Design and Engineering
• /
• v.12 no.4
• /
• pp.283-297
• /
• 2007

Seamless data integration in the CAx chain of the CAD/CAPP/CAM/CNC has been achieved to a high degree, but research concerning the transfer of data from conceptual sketches to a CAD system should be carried out further. This paper presents a method for reconstructing a 3D model from a freehand sketch. Sketch-based modeling research can be classified into gestural modeling methods and reconstructional modeling methods. This research involves the reconstructional modeling method. Here, Mitani's seminal work, designed for box-shaped 3D model using a predefined template, is improved by leveraging a relational template and specialized for automotive design. Matching between edge graphs of the relational template and the sketch is formulated and solved as the assignment problem using the feature vectors of the edges. Including the stroke preprocessing method required to generate an edge graph from a sketch, necessary procedures and relevant techniques for implementing the template-based modeling method are described. Examples from a working implementation are given.

• 한국HCI학회:학술대회논문집
• /
• 2006.02a
• /
• pp.1137-1142
• /
• 2006

3D 가상공간을 협업에 효과적으로 이용하기 위해서는 3D 가상공간에서 빠르고 쉽게 의사를 표현할 수 있는 기술이 필요하다. 본 연구에서는 이런 효과적인 의사표현 방법으로 Sketch 기법을 제안한다. Sketch 기법은 간단한 2D 선들의 표현을 통해 매우 빠르게 아이디어를 표현 할 수 있으며, 의도에 따라 부분을 강조하거나 가감할 수 있기 때문에 좀 더 직관적으로 의도를 전달 할 수 있다. 그래서 이러한 Sketch 를 공간상에서 손쉽게 3D 모델의 표면이나 공간 위에 입력하고, 입력한 Sketch 를 3D 가상공간과 연동하여 보여 줌으로써 쉽게 3D 가상공간 안에 Sketch 로 의견을 표현 할 수 있게 하는 Sketch 기반 인터페이스를 연구 하였다. Sketch 기법을 3D 가상공간에 적용하기 위해서는 3D 가상공간 상에서 2D Sketch 를 입력하기 위한 방법과 입력 시 발생할 수 있는 위상차의 문제를 해결해야 한다. 이 연구에서는 2D Sketch 입력을 대상 3D 객체 중심에서 Sketch할 부분의 Sketch plain 을 선택 및 생성함으로써 입력하는 방법을 연구하였으며, 입력 시 발생하는 위상차는 View-point 및 View-Plain 의 이동 등을 통해 해결점을 찾고자 하였다. 그리고 추후 연구를 통해 Sketch 를 Annotation 으로 활용, 협업에서 필요로 하는 Sketch Annotation 으로 개발하고자 한다.

• Journal of Internet Computing and Services
• /
• v.21 no.1
• /
• pp.33-43
• /
• 2020

Retrieving a 3D model from a 3D database and augmenting the retrieved model in the Augmented Reality system simultaneously became an issue in developing the plausible AR environments in a convenient fashion. It is considered that the sketch-based 3D object retrieval is an intuitive way for searching 3D objects based on human-drawn sketches as query. In this paper, we propose a novel deep learning based approach of retrieving a sketch-based 3D object as for an Augmented Reality Model. For this work, we introduce a new method which uses Sketch CNN, Wasserstein CNN and Wasserstein center loss for retrieving a sketch-based 3D object. Especially, Wasserstein center loss is used for learning the center of each object category and reducing the Wasserstein distance between center and features of the same category. The proposed 3D object retrieval and augmentation consist of three major steps as follows. Firstly, Wasserstein CNN extracts 2D images taken from various directions of 3D object using CNN, and extracts features of 3D data by computing the Wasserstein barycenters of features of each image. Secondly, the features of the sketch are extracted using a separate Sketch CNN. Finally, we adopt sketch-based object matching method to localize the natural marker of the images to register a 3D virtual object in AR system. Using the detected marker, the retrieved 3D virtual object is augmented in AR system automatically. By the experiments, we prove that the proposed method is efficiency for retrieving and augmenting objects.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.209-220
• /
• 2024

A sketch is an intuitive means to express information, but compared to actual images, it has the problem of being highly abstract, diverse, and sparse. Recent advances in deep learning models have made it possible to discover features that are common to images and sketches. In this paper, we summarize recent trends in sketch-based image retrieval (SBIR) but it is not limited to SBIR. Besides SBIR, we also introduce sketch-based image recognition and generation studies. Zero-shot learning enables models to recognize categories not encountered during training. Zero-shot SBIR methods are also discussed. Commonly used free-hand sketch datasets are summarized and retrieval performance based on these datasets is reported.

• Korean Journal of Computational Design and Engineering
• /
• v.13 no.4
• /
• pp.255-264
• /
• 2008

Research that reconstructs a 3D model from a freehand 2D sketch has gained attention since 1990s, when data integration in the CAD/CAPP/CAM/CNC chain was an important issue. However, 2D sketches in the conceptual design phase have not been integrated with the downstream CAD/CAPP/CAM/CNC chain. In this paper, we present a method to interface a freehand sketch modeling to commercial CAD systems by mapping a sketch modeling history to the macro parametric history. We use an extended ISO10303-112 standard to represent the modeling history in a gestural modeling system and translate sketch files to neutral macro files. Macro parametric translators are used to translate netural macro files to commercial CAD files.

• Journal of Korea Multimedia Society
• /
• v.12 no.10
• /
• pp.1466-1477
• /
• 2009

With development of 3D virtual space technology, diverse studies on 3D virtual space are being conducted because virtual space is being recognized as an appropriate technology for the field of urban and architectural design. However, while the process of urban and architectural design is done based on collaboration of various interested parties, there is lack of studies that support such collaboration. Communication within the virtual space must be provided for 3D virtual space to support collaboration, and a quick and direct technique of expressing one's intentions is required. sketch technique is effective. Ambiguous lines of sketch are extremely effective in that the range of thoughts in the idea stage can be broadened. However, such traditional sketch technique alone is difficult to support collaboration in 3D virtual space. That is because diverse shapes of 3D space require input of sketch. Accordingly in this study, sketch annotation that can effectively deliver intentions in 3D space was studied. Annotation technique using sketch-box was examined.

• International Journal of CAD/CAM
• /
• v.7 no.1
• /
• pp.113-120
• /
• 2007

Sketch based query plays an important role in shape matching for 3D shape retrieval system. Some researchers suggest the sketch based query interfaces. It is more effective to capture the needs of users rather than query by example. In this paper, we propose a new 2D sketch based query interface with coarse filters for shape matching. Coarse filtering enables to eliminate unfavorable shapes from shapes in DB. For the purpose of coarse filtering, we use two filters with the topological and geometric patterns. For the validity of our method, we show the experimental results.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.4 no.2 s.7
• /
• pp.45-52
• /
• 2005

Sketch-Planning based ITS Deployment Analysis System is an analysis tool for determining the impacts, benefits, and costs of various intelligent transportation system (ITS) deployments. This methodology offers one tool that systematically assesses ITS scenarios to facilitate the integration of such scenarios into the ongoing transportation planning process. Sketch-planting based methodology can estimate relative impacts, benefits, and costs for more than 60 types of ITS investments deployed in isolation or in any combination defined by the user.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.11a
• /
• pp.623-625
• /
• 2012

In this paper, we present a method to automatically generate a color pencil sketch image from a photo. First the image is converted into a sketch using a gradient estimation and then the color pencil sketch is produced by linear interpolation with original image and the sketched image. The experimental results show that the final image has a visual aspect of a color pencil sketch like image.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.1268-1271
• /
• 2022

Since it is common to have touch-screen devices, it is less challenging to draw sketches anywhere and save them in vector form. Current research on sketches considers coordinate sequence data and adopts sequential models for learning sketch representation in sketch understanding. In the sketch dataset, it has become customary that the dataset is in vector coordinate format. Moreover, the popular dataset does not consider real-life sketches, sketches from pencil, pen, and paper. Art psychology uses real-life sketches to analyze patients. ETRI presents a unique sketch dataset for sketch recognition of autism spectrum disorder in pixel format. We present a method to formulate the dataset for better generalization of sketch data. Through experiments, we show that pixel-based models can produce a good performance.