Masked Autoencoder (MAE) has shown remarkable potential in self-supervised representation learning for 3D point clouds. However, these methods primarily rely on point-level or low-level feature ...

Hi! I'm simply looking to get access to the masked autoencoder weights used to train SAM2 for usage in a few different projects, including attempting to expand upon Voxel51's recent research with a ...

The traditional masked reconstruction model uses the autoencoder structure but cannot learn richer potential information and data structure. To this end, we propose to improve the latent space based ...

Improve this page Add a description, image, and links to the masked-autoencoder topic page so that developers can more easily learn about it.

Masked autoencoder (MAE) is a recently widely used self-supervised learning method that has achieved great success in NLP and computer vision. However, the potential advantages of masked pre-training ...