[2018.04.11] Multimodal learning for Audio-Visual Scene

[unizard]

Audio-Visual Scene Understanding

• These days, multi-modal learning is a trend. However, I do not yet catch up the reason why this item is booming. (what is a technical breakthrough?) Anyway, we need to read following works ASAP.

  Lip Reading, ACCV2016, CVPR2017 Visually Indicated Sounds, NIPS2016 Learning to Localize Sound Source in Visual Scenes, CVPR2018 The Sound of Pixels, ECCV2018 Audio-Visual Scene Analysis with Self-Supervised Multisensory Features, ECCV2018