search

Justin1904 / Low-rank-Multimodal-Fusion

This is the repository for "Efficient Low-rank Multimodal Fusion with Modality-Specific Factors", Liu and Shen, et. al. ACL 2018
254 stars 47 forks source link

about the original face features and audio features #4

Closed hugddygff closed 5 years ago

hugddygff commented 5 years ago

Hi, Justin, Can you share the original face features and audio features with averaging along the temporal dimension? and where can I download the original video data of the three datasets?

Thanks very much!

Justin1904 commented 5 years ago

The link in the README should point to the visual and acoustic features averaged across the temporal dimension. For the original video data, you can download them here. For more info on these datasets, you can check out CMU-MultimodalSDK.

hugddygff commented 5 years ago

sorry, I have made a spelling mistake in the question. I mean the original features without averaging along the temporal dimension. Instead of with， thanks！

Justin1904 commented 5 years ago

For that, you could also refer to the Multimodal SDK mentioned above.