Learning Deep Features for Discriminative Localization

CVPR 2016 Bolei ZhouAditya KhoslaAgata LapedrizaAude OlivaAntonio Torralba

In this work, we revisit the global average pooling layer proposed in [13], and shed light on how it explicitly enables the convolutional neural network to have remarkable localization ability despite being trained on image-level labels. While this technique was previously proposed as a means for regularizing training, we find that it actually builds a generic localizable deep representation that can be applied to a variety of tasks... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Weakly-Supervised Object Localization ILSVRC 2015 AlexNet-GAP Top-1 Error Rate 67.19 # 2
Weakly-Supervised Object Localization ILSVRC 2016 VGGnet-GAP Top-5 Error 45.14 # 3
Weakly-Supervised Object Localization ILSVRC 2016 AlexNet-GAP Top-5 Error 52.16 # 4
Weakly-Supervised Object Localization Tiny ImageNet CAM Top-1 Localization Accuracy 40.55 # 2

Methods used in the Paper