Transfer Learning for Rodent Behavior Recognition

Many behavior recognition systems are trained and tested on single datasets limiting their application to comparable datasets. While retraining the system with a novel dataset is possible, it involves laborious annotation effort. We propose to minimize the annotation effort by reusing the knowledge obtained from previous datasets and adapting the recognition system to the novel data. To this end, we investigate the use of transfer learning in the context of rodent behavior recognition. Specifically, we look at two transfer learning methods with two different approaches and examine the implications of their respective assumptions on synthetic data. We further illustrate their performance in transferring a rat action classifier to a mouse action classifier. The performance results in the transfer task are promising. The classification accuracy improves substantially with only very few labeled examples from the novel dataset.