Places-CNDS-8 is a "8conv3fc layer" deep Convolutional neural Networks model trained on MIT Places Dataset with Deep Supervision.
Top-1 and Top-5 accuracies (%) on MIT Places Val/Test dataset:
top-1 val/test top-5 val/test
54.7/55.7 84.1/85.8
The method is described in the following report. Thanks for citing this report if the model is useful for your research.
Downloading our 8conv3fc_CNDS.caffemodel (8 convolutional layers and 3 fully connected layers)
https://drive.google.com/file/d/0B_s0NuQ5DMW_SUZCOUV2amY3ZkU/view?usp=sharing
Thanks for citing our work if the model is useful for you.
http://arxiv.org/pdf/1505.02496.pdf
Training Deeper Convolutional Networks with Deep Supervision
L.Wang, C.Lee, Z.Tu, S. Lazebnik, arXiv:1505.02496, 2015