Pre-trained Models

We provide a number of pre-trained models for easy use. When using these models, please cite and heed licenses accordingly! We also welcome community contributions via pull requests to the repository.

You can find a number of demos showing how to download and run the models below here.

The List

This list contains links to architecture definitions (*.json), download scripts for the mean images (*.tensor), and pre-trained weights (*.marvin). We have also taken care to give credit where it is due, so please let us know if there is anything wrong or missing.

Architecture Mean image Weights Trainer Citation License
AlexNet for ImageNet ImageNet1k Download BVLC C1 C2 C3 L1 L2 L3
AlexNet for Places Places205 Download Places C2 C4 L2 L4
GoogLeNet for ImageNet ImageNet1k Download BVLC C1 C3 C5 L1 L3 L5
GoogLeNet for Places Places205 Download Places C4 C5 L4 L5
VGGNet 16 for ImageNet ImageNet1k Download VGG C3 C6 L3 L6
VGGNet 19 for ImageNet ImageNet1k Download VGG C3 C6 L3 L6

Citations

  1. Jia, Yangqing, et al. "Caffe: Convolutional architecture for fast feature embedding." Proceedings of the ACM International Conference on Multimedia. 2014.
  2. Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks." Advances in neural information processing systems. 2012.
  3. Deng, Jia, et al. "Imagenet: A large-scale hierarchical image database." IEEE Conference on Computer Vision and Pattern Recognition. 2009.
  4. Zhou, Bolei, et al. "Learning deep features for scene recognition using places database." Advances in Neural Information Processing Systems. 2014.
  5. Szegedy, Christian, et al. "Going deeper with convolutions." arXiv preprint arXiv:1409.4842. 2014.
  6. Simonyan, Karen, and Andrew Zisserman. "Very deep convolutional networks for large-scale image recognition." arXiv preprint arXiv:1409.1556. 2014.
  7. </ol>

    Licenses

    1. License here
    2. Released for unrestricted use (link)
    3. Depending on your use case, custom license (link)
    4. Original annotations under Creative Commons (link)
    5. Released for unrestricted use (link)
    6. Released under Creative Commons 4.0 (link) </ol>