Large-Scale Knowlege Transfer

[Full set of software and data is available upon request]


Semantic relatedness from Language resource

For now see our previous release at We will shortly add an updated version.

Large Scale Learning

In our recent work [1] we learn a SVM with MeanSGD. The Matlab/C++ trains several classes in parallel using multi-core hardware, loading the data only once to memory. This software is based on code by Leon Bottou.

meanSGD 1.1

Change log

1.1: Added input check and "-largeArrayDims" for mex compilation.


Undefined function or variable 'meanSvmsgd2HingelossOpenmp'

Make sure the binary is on the path and you provide data of the type "single". Alternatively change the C++ code to double.

Image Features

The features are zipped Matlab (mat) files, one per image.

Precomputed SVM classifier scores

Groundtruth labels

Known Issues

  • black-white images as we use rgSift.



Help & Contact

If you need any help or have any suggestions feel free to contact me.


[1] Evaluating Knowledge Transfer and Zero-Shot Learning in a Large-Scale Setting, M. Rohrbach, M. Stark and B. Schiele, IEEE Conference on Computer Vision and Pattern Recognition, June, (2011)