search

MCG-NJU / VideoMAE

[NeurIPS 2022 Spotlight] VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training
https://arxiv.org/abs/2203.12602
Other
1.32k stars 131 forks source link

learnable position embedding #83

Closed kun-dragon closed 1 year ago

kun-dragon commented 1 year ago

Do you try to use learnable position embedding, instead of sinusoid encoding? Will it have better performance?

yztongzhan commented 1 year ago

We follow MAE and use sinusoid encoding as default setting.