Skip to Main content Skip to Navigation
Conference papers

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

Maxime Oquab 1, 2, 3 Léon Bottou 4 Ivan Laptev 2, 3 Josef Sivic 2, 3
3 WILLOW - Models of visual object recognition and scene understanding
CNRS - Centre National de la Recherche Scientifique : UMR8548, Inria Paris-Rocquencourt, DI-ENS - Département d'informatique de l'École normale supérieure
Abstract : 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 prediction 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.
Document type :
Conference papers
Complete list of metadata

Cited literature [61 references]  Display  Hide  Download
Contributor : Josef Sivic Connect in order to contact the contributor
Submitted on : Sunday, May 17, 2015 - 9:56:24 PM
Last modification on : Wednesday, November 17, 2021 - 12:31:47 PM
Long-term archiving on: : Tuesday, September 15, 2015 - 1:12:43 AM


Files produced by the author(s)


  • HAL Id : hal-01015140, version 2



Maxime Oquab, Léon Bottou, Ivan Laptev, Josef Sivic. Is object localization for free? – Weakly-supervised learning with convolutional neural networks. IEEE Conference on Computer Vision and Pattern Recognition, Jun 2015, Boston, United States. ⟨hal-01015140v2⟩



Record views


Files downloads