Active learning based clothing image recommendation with implicit user preferences

Chiao-Meng Huang, Chia-Po Wei, Yu-Chiang Frank Wang

Abstract: We address the problem of user-specific clothing image recommendation in this paper. Different from prior retrieval approaches, we advance an active learning scheme during retrieval for inferring user preferences. With a recently developed sparse-coding based algorithm for content-based image retrieval, we utilize support vector regression (SVR) with a user-interaction training stage to observe user preferences based on the feedback of retrieval results. Therefore, there is no need to explicitly ask his/her preferences such as desirable colors or patterns of clothing images. A subjective evaluation on a commercial clothing image dataset confirms the effectiveness of our method, which is shown to produce more satisfactory recommendation results when comparing to state-of-the-art content-based image retrieval approaches.