xiaomengyc / SPG

(ECCV2018) Self-produced Guidance for Weakly-supervised Object Localization
https://arxiv.org/abs/1807.08902
MIT License
146 stars 25 forks source link

backbone architecture? #7

Open d-li14 opened 5 years ago

d-li14 commented 5 years ago

@xiaomengyc Hi, the whole framework adopts modified Inception v3 as the backbone, but why the weights of pre-trained VGG-16 is restored in the training command line here? I suppose the correct model path is "~/.torch/models/inception_v3_google-1a9a5a14.pth"?