Selective visual attention enables learning and recognition of multiple objects in cluttered scenes Export

@article{citeulike:463067, abstract = {A key problem in learning representations of multiple objects from unlabeled images is that it is a priori impossible to tell which part of the image corresponds to each individual object, and which part is irrelevant clutter. Distinguishing individual objects in a scene would allow unsupervised learning of multiple objects from unlabeled images. There is psychophysical and neurophysiological evidence that the brain employs visual attention to select relevant parts of the image and to serialize the perception of individual objects. We propose a method for the selection of salient regions likely to contain objects, based on bottom-up visual attention. By comparing the performance of David Lowe's recognition algorithm with and without attention, we demonstrate in our experiments that the proposed approach can enable one-shot learning of multiple objects from complex scenes, and that it can strongly improve learning and recognition performance in the presence of large amounts of clutter.}, author = {Walther, Dirk and Rutishauser, Ueli and Koch, Christof and Perona, Pietro}, citeulike-article-id = {463067}, citeulike-linkout-0 = {http://dx.doi.org/10.1016/j.cviu.2004.09.004}, citeulike-linkout-1 = {http://www.sciencedirect.com/science/article/B6WCX-4GDC091-1/2/917ef04c80e6be35c25c6dc9543b65f0}, doi = {10.1016/j.cviu.2004.09.004}, journal = {Computer Vision and Image Understanding}, keywords = {attention, sift, vision}, month = {}, number = {1-2}, pages = {41--63}, posted-at = {2006-01-12 14:51:41}, priority = {0}, title = {Selective visual attention enables learning and recognition of multiple objects in cluttered scenes}, url = {http://dx.doi.org/10.1016/j.cviu.2004.09.004}, volume = {100}, year = {2005} }

TY - JOUR ID - citeulike:463067 L3 - citeulike-article-id:463067 N2 - A key problem in learning representations of multiple objects from unlabeled images is that it is a priori impossible to tell which part of the image corresponds to each individual object, and which part is irrelevant clutter. Distinguishing individual objects in a scene would allow unsupervised learning of multiple objects from unlabeled images. There is psychophysical and neurophysiological evidence that the brain employs visual attention to select relevant parts of the image and to serialize the perception of individual objects. We propose a method for the selection of salient regions likely to contain objects, based on bottom-up visual attention. By comparing the performance of David Lowe's recognition algorithm with and without attention, we demonstrate in our experiments that the proposed approach can enable one-shot learning of multiple objects from complex scenes, and that it can strongly improve learning and recognition performance in the presence of large amounts of clutter. IS - 1-2 JF - Computer Vision and Image Understanding EP - 63 TI - Selective visual attention enables learning and recognition of multiple objects in cluttered scenes VL - 100 SP - 41 KW - attention KW - sift KW - vision AU - Walther, Dirk AU - Rutishauser, Ueli AU - Koch, Christof AU - Perona, Pietro PY - 2005/0// UR - http://dx.doi.org/10.1016/j.cviu.2004.09.004 ER -