Video dataset should contain a single person with moving face gestures so that the trained model does not depend on a single pose. I want to train facenet VGG face retrained models on this dataset using triplet loss for transfer learning .
Mouaaz Ranjha
These links might help, have a look:
1. https://github.com/topics/triplet-loss
2. https://awesomeopensource.com/projects/triplet-loss
DEAR Mouaaz Ranjha
https://arxiv.org/pdf/2005.00355.pdf
https://ieeexplore.ieee.org/document/9191289
20BN-SOMETHING-SOMETHING. The 20BN-SOMETHING-SOMETHING dataset is a large collection of densly-labeled video clips that show humans performing predefined basic actions with every day objects. Human activities
https://github.com/topics/triplet-loss
https://arxiv.org/pdf/1409.4127.pdf
https://arxiv.org/pdf/1409.4127.pdf
- Badges
- Science topic