This paper describes a multi-pitch analysis method using specmurt analysis with iterative estimation of the quasioptimal common harmonic structure function. Specmurt analysis (Sag...
We introduce a new model for extracting classified structural segments, such as intro, verse, chorus, break and so forth, from recorded music. Our approach is to classify signal ...
Samer A. Abdallah, Katy Noland, Mark B. Sandler, M...
This paper proposes a content-based medical image retrieval (CBMIR) framework using dynamically optimized features from multiple regions of medical images. These regional features...
Wei Xiong, Bo Qiu, Qi Tian, Changsheng Xu, Sim Hen...
Determining similarity is a fundamental task in querying multimedia databases in a content-based way. For this challenging task, there exist numerous similarity models which measu...
Robust semantic labeling of image regions is a basic problem in representing and retrieving image/video content. We propose an SVM-MRF framework to model features and their spatia...