Andre Holzapfel, from INESC Porto, gives a talk entitled "Selective sampling for beat tracking evaluation" on Friday March 2nd at 4pm in room 55.410.

Abstract: An approach is presented that identifies music samples which are difficult for current state-of-the-art beat trackers. In order to estimate this difficulty even for examples without ground truth, a method motivated by selective sampling is applied. This method assigns a degree of difficulty to a sample based on the mutual dis-agreement between the output of various beat tracking systems. On a large beat annotated dataset we show that this mutual agreement is correlated with the mean performance of the beat trackers evaluated against the ground truth, and hence can be used to identify difficult examples by predicting poor beat tracking performance. Towards the aim of advancing future beat tracking systems, we form a new dataset containing a high proportion of challenging music examples based on our method. We analyze the relations between perceptual difficulty and difficulty for automatic beat tracking using these data, and propose which signal properties are characterized by the highest potential of improvement of automatic beat tracking.

Andre Holzapfel webpage