The compressed_dim parameter for cnn features

martin-danelljan / Continuous-ConvOp

The Continuous Convolution Operator Tracker (C-COT).

GNU General Public License v3.0

196 stars 83 forks source link

The compressed_dim parameter for cnn features #5

Closed egundogdu closed 7 years ago

egundogdu commented 7 years ago

Hi Martin, In the VOT2016_settings of CCOT, the following line exists. cnn_params.compressed_dim = [16 64];
Nevertheless, I could not find its usage to compressed the feature maps to the values in the RHS of the above command. Could you please clarify this point ?

martin-danelljan commented 7 years ago

Hi. Thank you for your comment. Those parameters are not used in C-COT and were accidentally included while copy-pasting code. They are used in our new ECO tracker though to set the dimensionality of the compressed features.