Conference Paper

Generative Object Definition and Semantic Recognition.

DOI: 10.2312/3DOR/3DOR11/001-008 Conference: Eurographics Workshop on 3D Object Retrieval 2011, Llandudno, UK, April 10, 2011. Proceedings
Source: DBLP


''What is the difference between a cup and a door?'' These kinds of questions have to be answered in the context of digital libraries. This semantic information, which describes an object on a high, abstract level, is needed in order to provide digital library services such as indexing, markup and retrieval. In this paper we present a new approach to encode and to extract such semantic information. We use generative modeling techniques to describe a class of objects: each class is represented by one algorithm; and each object is one set of high-level parameters, which reproduces the object if passed to the algorithm. Furthermore, the algorithm is annotated with semantic information, i.e. a human-readable description of the object class it represents. We use such an object description to recognize objects in real-world data e.g. laser scans. Using an algorithmic object description, we are able to identify 3D subparts, which can be described and generated by the algorithm. Furthermore, we can determine the needed input parameters. In this way, we can classify objects, recognize them semantically and we can determine their parameters (cup's height, radius, etc.).

1 Follower
11 Reads
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: In the context of computer graphics, a generative model is the description of a three-dimensional shape: Each class of objects is represented by one algorithm $M$. Furthermore, each described object is a set of high-level parameters $x$, which reproduces the object, if an interpreter evaluates $M(x)$. This procedural knowledge differs from other kinds of knowledge, such as declarative knowledge, in a significant way. Generative models are designed by programming. In order to make generative modeling accessible to non-computer scientists, we created a generative modeling framework based on the easy-to-use scripting language JavaScript (JS). Furthermore, we did not implement yet another interpreter, but a JS-translator and compiler. As a consequence, our framework can translate generative models from JavaScript to various platforms. In this paper we present an overview of Euclides and quintessential examples of supported platforms: Java, Differential Java, and GML. Java is a target language, because all frontend and framework components are written in Java making it easier to be embedded in an integrated development environment. The Differential Java backend can compute derivatives of functions, which is a necessary task in many applications of scientific computing, e.g., validating reconstruction and fitting results of laser scanned surfaces. The postfix notation of GML is very similar to that of Adobes Postscript. It allows the creation of high-level shape operators from low-level shape operators. The GML serves as a platform for a number of applications because it is extensible and comes with an integrated visualization engine. This innovative meta-modeler concept allows a user to export generative models to other platforms without losing its main feature – the procedural paradigm. In contrast to other modelers, the source code does not need to be interpreted or unfolded, it is translated. Therefore, it can still be a very compact representation of a complex model.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The paper presents new techniques and processing results for automatic segmentation, shape classification, generic pose estimation, and model-based identification of naval vessels in laser radar imagery. The special characteristics of focal plane array laser radar systems such as multiple reflections and intensity-dependent range measurements are incorporated into the algorithms. The proposed 3D model matching technique is probabilistic, based on the range error distribution, correspondence errors, the detection probability of potentially visible model points and false alarm errors. The match algorithm is robust against incomplete and inaccurate models, each model having been generated semi-automatically from a single range image. A classification accuracy of about 96% was attained, using a maritime database with over 8000 flash laser radar images of 146 ships at various ranges and orientations together with a model library of 46 vessels. Applications include military maritime reconnaissance, coastal surveillance, harbor security and anti-piracy operations.
    Proceedings of SPIE - The International Society for Optical Engineering 05/2012; 8391:9-. DOI:10.1117/12.920264 · 0.20 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: A procedural model is a script, which generates a geometric object. The script's input parameters offer a simple way to specify and modify the scripting output. Due to its algorithmic character, a procedural model is perfectly suited to describe geometric shapes with well-organized structures and repetitive forms. In this paper, we interpret a generative script as a function, which is nested into an objective function. Thus, the script's parameters can be optimized according to an objective. We demonstrate this approach using architectural examples: each generative script creates a building with several free parameters.The objective function is an energy-efficiency-simulation that approximates a building's annual energy consumption. Consequently, the nested objective function reads a set of building parameters and returns the energy needs for the corresponding building. This nested function is passed to a minimization and optimization process. Outcome is the best building (within the family of buildings described by its script) concerning energy efficiency. Our contribution is a new way of modeling. The generative approach separates design and engineering: the complete design is encoded in a script and the script ensures that all parameter combinations (within a fixed range) generate a valid design. Then the design can be optimized numerically.
    01/2013; 2:123-128. DOI:10.1109/IECON.2013.6699904
Show more