Audio-Visual Embedding for Cross-Modal MusicVideo Retrieval through Supervised Deep CCA [article]

Donghuo Zeng, Yi Yu, Keizo Oyama
2019 arXiv   pre-print
Deep learning has successfully shown excellent performance in learning joint representations between different data modalities. Unfortunately, little research focuses on cross-modal correlation learning where temporal structures of different data modalities, such as audio and video, should be taken into account. Music video retrieval by given musical audio is a natural way to search and interact with music contents. In this work, we study cross-modal music video retrieval in terms of emotion
more » ... ilarity. Particularly, audio of an arbitrary length is used to retrieve a longer or full-length music video. To this end, we propose a novel audio-visual embedding algorithm by Supervised Deep CanonicalCorrelation Analysis (S-DCCA) that projects audio and video into a shared space to bridge the semantic gap between audio and video. This also preserves the similarity between audio and visual contents from different videos with the same class label and the temporal structure. The contribution of our approach is mainly manifested in the two aspects: i) We propose to select top k audio chunks by attention-based Long Short-Term Memory (LSTM)model, which can represent good audio summarization with local properties. ii) We propose an end-to-end deep model for cross-modal audio-visual learning where S-DCCA is trained to learn the semantic correlation between audio and visual modalities. Due to the lack of music video dataset, we construct 10K music video dataset from YouTube 8M dataset. Some promising results such as MAP and precision-recall show that our proposed model can be applied to music video retrieval.
arXiv:1908.03744v1 fatcat:2l2vdm7a7zatvdbk6ja2tfdmqi