search

facebookresearch / segment-anything

The repository provides code for running inference with the SegmentAnything Model (SAM), links for downloading the trained model checkpoints, and example notebooks that show how to use the model.
Apache License 2.0
46.55k stars 5.52k forks source link

MAE Pretrained Encoder #249

Open HighwayWu opened 1 year ago

HighwayWu commented 1 year ago

Thanks for your awesome work. If possible, could the pretrained MAE or ImageNet weights on the SAM encoder (ViT with input size 1024) be released? Since the encoder is different from MAE or ImageNet pretrained ViT that has input size 224. Thanks again.

simoneangarano commented 1 year ago

Same request here.

ByungKwanLee commented 10 months ago

https://github.com/ByungKwanLee/Full-Segment-Anything addresses the ciritical issues of SAM, which supports batch-input on the full-grid prompt (automatic mask generation) with post-processing: removing duplicated or small regions and holes, under flexible input image size

betterze commented 8 months ago

I have similar request. Could you release the pretrained mask autoencoder weights on 1024x1024 resolution? thx a lot

Syloveslife commented 7 months ago

Similar request.