Fast concurrent object localization and recognition

Object localization and classification are important problems in computer vision. However, in many applications, exhaustive search over all class labels and image locations is computationally prohibitive. While several methods have been proposed to make either classification or localization more efficient, few have dealt with both tasks simultaneously. This paper proposes an efficient method for concurrent object localization and classification based on a data-dependent multi-class branch-and-bound formalism. Existing bag-of-features classification schemes, which can be expressed as weighted combinations of feature counts can be readily adapted to our method. We present experimental results that demonstrate themerit of our algorithmin terms of classification accuracy, localization accuracy, and speed, compared to baseline approaches including exhaustive search, the ISM method, and single-class branch and bound.

• Type: Conference
• Year: 2009
• Where: CVPR
• Authors: Tom Yeh, John J. Lee, Trevor Darrell
• Keywords: Computer Vision, CVPR 2009, Object Classification, Object Clocalization