Biased support vector machine active learning for 3D model retrieval

Relevance feedback is an iterative search technique to bridge the semantic gap between the high level user intention and low level data representation. This technique interactively determines a user's desired output or query concept by asking the user whether certain proposed 3D models are relevant or not. For a relevance feedback algorithm to be effective, it must grasp a user's query concept accurately and quickly. In this paper, we propose the use of biased support vector machine active learner for conducting relevance feedback for 3D model retrieval. The algorithm selects the most informative 3D models to query a user and quickly learns a boundary that separates the 3D models that satisfy the user's query concept from the rest of the dataset. Experimental results show that this algorithm achieves higher search accuracy than traditional query refinement schemes.