Modeling Sensor Performance for Model-Based Vision

Abstract

The model-based vision requires object appearances in the computer. How an object appears in the image is a result of interaction between the object properties and the sensor characteristics. Thus, in model-based vision, we ought to model the sensor as well as modeling the object. In the past, however, the sensor model was not used in the model-based vision or, at least, was contained in the object model implicitly. This paper presents a framework between an object model and the object appearances. We consider two aspects of sensor characteristics: sensor detectability and sensor reliability. Sensor detectability specifies what kind of features can be detected and in what area the features are detected; sensor reliability specifies how reliable detected features are. Commonly available sensors are briefly examined in terms of their sensor characteristics. We define the configuration space to represent sensor characteristics. We propose a representation method of the sensor detectability in the configuration space. Sensor reliability distribution is also discussed in the configuration space. Under this framework. we characterize the photometric stereo and the light-stripe range finder as examples.