Is object localization for free? - Weakly-supervised learning with convolutional neural networks

Maxime Oquab, Leon Bottou, Ivan Laptev, Josef Sivic
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Successful methods for visual object recognition typically rely on training datasets containing lots of richly annotated images. Detailed image annotation, e.g. by object bounding boxes, however, is both expensive and often subjective. We describe a weakly supervised convolutional neural network (CNN) for object classification that relies only on image-level labels, yet can learn from cluttered scenes containing multiple objects. We quantify its object classification and object location
more » ... on performance on the Pascal VOC 2012 (20 object classes) and the much larger Microsoft COCO (80 object classes) datasets. We find that the network (i) outputs accurate image-level labels, (ii) predicts approximate locations (but not extents) of objects, and (iii) performs comparably to its fully-supervised counterparts using object bounding box annotation for training.
doi:10.1109/cvpr.2015.7298668 dblp:conf/cvpr/OquabBLS15 fatcat:ivhq7y7465gbhlj6ouaqw5sk3i