maskbit
Collection
6 items
•
Updated
This model is the MaskBit tokenizer with a vocabulary size of 14bits. It uses a downsampling factor of 16 and is trained on ImageNet for images of resolution 256.
You can find more details on the project page and in the paper.
Inference Providers
NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API:
The model has no library tag.
Dataset used to train markweber/maskbit_tokenizer_14bit
Collection including markweber/maskbit_tokenizer_14bit
Evaluation results
- rFID on ILSVRC/imagenet-1kself-reported1.370
- InceptionScore on ILSVRC/imagenet-1kself-reported190.300
- LPIPS on ILSVRC/imagenet-1kself-reported0.286
- PSNR on ILSVRC/imagenet-1kself-reported21.500
- SSIM on ILSVRC/imagenet-1kself-reported0.560
- CodebookUsage on ILSVRC/imagenet-1kself-reported1.000