Browse by topic
Type of publication
Local Representations for Multi-Object Recognition. In DAGM 2003, Pattern Recognition, 25th DAGM Symposium, 2003.We present a method using local representations to recognize multiple objects in images with and without background. Starting from a straight forward approach, we combine the use of local representations with segmentation and template matching to improve the results. The performance of the classifiers is evaluated on four image databases of different difficulties. All databases consist of images containing one, two or three objects and differ in the backgrounds which are used and the occurrence of occlusions. Classification results are promising considering the difficulty of the task.