How did you extract your feature of video？

LiJiaBei-7 / rivrl

Source code of our TCSVT'22 paper Reading-strategy Inspired Visual Representation Learning for Text-to-Video Retrieval

Apache License 2.0

19 stars 1 forks source link

Open Siaaaaaa1 opened 1 month ago

Siaaaaaa1 commented 1 month ago

How did you extract your feature of video？I'm going to use new videos in your model，but how can i convert my video to feature and feed in your model？

LiJiaBei-7 commented 1 month ago

For video features, we follow the Dual Encoding, which you can find in https://github.com/danieljf24/hybrid_space