Title:Performance Evaluation in Multimedia Retrieval

Abstract:Performance evaluation in multimedia retrieval, as in the information retrieval domain at large, relies heavily on retrieval experiments, employing a broad range of techniques and metrics. These can involve human-in-the-loop and machine-only settings for the retrieval process itself and the subsequent verification of results. Such experiments can be elaborate and use-case-specific, which can make them difficult to compare or replicate. In this paper, we present a formal model to express all relevant aspects of such retrieval experiments, as well as a flexible open-source evaluation infrastructure that implements the model. These contributions intend to make a step towards lowering the hurdles for conducting retrieval experiments and improving their reproducibility.