Accurate Object Detection with Deformable Shape Models Learnt from Images Export

@inproceedings{FerrariCVPR2007, abstract = {We present an object class detection approach which fully integrates the complementary strengths offered by shape matchers. Like an object detector, it can learn class models directly from images, and localize novel instances in the presence of intra-class variations, clutter, and scale changes. Like a shape matcher, it finds the accurate boundaries of the objects, rather than just their bounding-boxes. This is made possible by 1) a novel technique for learning a shape model of an object class given images of example instances; 2) the combination of Hough-style voting with a non-rigid point matching algorithm to localize the model in cluttered images. As demonstrated by an extensive evaluation, our method can localize object boundaries accurately, while needing no segmented examples for training (only bounding-boxes).}, author = {Ferrari, V. and Jurie, F. and Schmid, C.}, booktitle = {Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on}, citeulike-article-id = {3544974}, citeulike-linkout-0 = {http://dx.doi.org/10.1109/CVPR.2007.383043}, citeulike-linkout-1 = {http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=4270068}, doi = {10.1109/CVPR.2007.383043}, journal = {Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on}, keywords = {cvpr, detection, learning, model, object, shape}, pages = {1--8}, posted-at = {2008-11-16 23:30:33}, priority = {2}, title = {Accurate Object Detection with Deformable Shape Models Learnt from Images}, url = {http://dx.doi.org/10.1109/CVPR.2007.383043}, year = {2007} }

TY - CONF ID - FerrariCVPR2007 L3 - citeulike-article-id:3544974 N2 - We present an object class detection approach which fully integrates the complementary strengths offered by shape matchers. Like an object detector, it can learn class models directly from images, and localize novel instances in the presence of intra-class variations, clutter, and scale changes. Like a shape matcher, it finds the accurate boundaries of the objects, rather than just their bounding-boxes. This is made possible by 1) a novel technique for learning a shape model of an object class given images of example instances; 2) the combination of Hough-style voting with a non-rigid point matching algorithm to localize the model in cluttered images. As demonstrated by an extensive evaluation, our method can localize object boundaries accurately, while needing no segmented examples for training (only bounding-boxes). JF - Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on EP - 8 TI - Accurate Object Detection with Deformable Shape Models Learnt from Images T2 - Computer Vision and Pattern Recognition, 2007. CVPR '07. IEEE Conference on SP - 1 KW - cvpr KW - detection KW - learning KW - model KW - object KW - shape AU - Ferrari, V AU - Jurie, F AU - Schmid, C PY - 2007/// UR - http://dx.doi.org/10.1109/CVPR.2007.383043 ER -