Background Filtering for Improving of Object Detection in Images
Authors:
- 1 August 2010
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10514651,p. 922-925
Abstract
We propose a method for improving object recognition in street scene images by identifying and filtering out background aspects. We analyse the semantic relationships between foreground and background objects and use the information obtained to remove areas of the image that are misclassified as foreground objects. We show that such background filtering improves the performance of four traditional object recognition methods by over 40%. Our method is independent of the recognition algorithms used for individual objects, and can be extended to generic object recognition in other environments by adapting other object models.Keywords
This publication has 10 references indexed in Scilit:
- A stochastic graph grammar for compositional object representation and recognitionPattern Recognition, 2009
- Object detection by global contour shapePattern Recognition, 2008
- Learning semantic object parts for object categorizationImage and Vision Computing, 2008
- LabelMe: A Database and Web-Based Tool for Image AnnotationInternational Journal of Computer Vision, 2007
- Robust Object Recognition with Cortex-Like MechanismsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2007
- Semantic Modeling of Natural Scenes for Content-Based Image RetrievalInternational Journal of Computer Vision, 2006
- Combining Top-Down and Bottom-Up Segmentation2005
- Dynamic learning from multiple examples for semantic object segmentation and searchComputer Vision and Image Understanding, 2004
- On internal representations in face recognition systemsPattern Recognition, 2000
- Efficient and effective Querying by Image ContentJournal of Intelligent Information Systems, 1994