Combining efficient object localization and image classification

In this paper we present a combined approach for ob- ject localization and classification. Our contribution is two- fold. (a) A contextual combination of localization and clas- sification which shows that classification can improve de- tection and vice versa. (b) An efficient two stage sliding window object localization method that combines the effi- ciency of a linear classifier with the robustness of a sophis- ticated non-linear one. Experimental results evaluate the parameters of our two stage sliding window approach and show that our combined object localization and classifica- tion methods outperform the state-of-the-art on the PASCAL VOC 2007 and 2008 datasets.

• Type: Conference
• Year: 2009
• Where: ICCV
• Authors: Hedi Harzallah, Frederic Jurie, Cordelia Schmid
• Keywords: Computer Vision, ICCV 2009, Image Classification, Object Localization