Disambiguation in Unknown Object Detection by Integrating Image and Speech Recognition Confidences
This paper presents a new method to detect unknown objects and their unknown names in object manipulation through man-robot dialog. In the method, the detection is carried out by using the information of object images and user’s speech in an integrated way. Originality of the method is to use logistic regression for the discrimination between unknown and known objects. The accuracy of the unknown object detection was 97% in the case when there were about fifty known objects.
KeywordsObject Recognition Speech Recognition Image Recognition Object Manipulation Speech Feature
Unable to display preview. Download preview PDF.
- 1.Araki, T., et al.: Autonomous Acquisition of Multimodal Information for Online Object Concept Formation by Robots. In: IEEE International Conference on Intelligent Robots and Systems (2011)Google Scholar
- 3.Nakano, M., et al.: Grounding New Words on The Physical World in Multi-Domain Human-Robot Dialogues. In: Dialog with Robots: Papers from the AAAI Fall Symposium (2010)Google Scholar
- 5.Skocaj, D., et al.: A basic cognitive system for interactive continuous learning of visual concepts. In: ICRA 2010 Workshop (2010)Google Scholar
- 6.Zuo, X., et al.: Detecting Robot-Directed Speech by Situated Understanding in Physical Interaction. Journal of Artificial Intelligence 25(25), 670–682 (2010)Google Scholar
- 7.Julius, http://julius.sourceforge.jp/
- 10.Kurita, T.: Interactive Weighted Least Squares Algorithms for Neural Networks Classifiers. In: Proc. Workshop on Algorithmic Learning Theory, pp. 77–86 (1992)Google Scholar