File size: 3,205 Bytes
3e539bc d04d08f 149b569 3e539bc 39bbff9 3e539bc d04d08f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 |
---
{}
---
# AM-RADIO: Reduce All Domains Into One
Mike Ranzinger, Greg Heinrich, Jan Kautz, Pavlo Molchanov
[NVIDIA Research](https://www.nvidia.com/en-us/research/)
\[[AM-RADIO Paper](https://arxiv.org/abs/2312.06709)\]
\[[PHI-S Paper](https://arxiv.org/abs/2410.01680)\]
\[[BibTex](#citing-radio)\]\[[GitHub examples](https://github.com/NVlabs/RADIO)\]
\[[Tech report on v2.5](https://github.com/NVlabs/RADIO/blob/main/RADIOv2.5_tech_report.md)\]
### HuggingFace Hub
You can pull the model from a Python script:
```Python
import torch
from PIL import Image
from transformers import AutoModel, CLIPImageProcessor
hf_repo = "nvidia/RADIO-B"
image_processor = CLIPImageProcessor.from_pretrained(hf_repo)
model = AutoModel.from_pretrained(hf_repo, trust_remote_code=True)
model.eval().cuda()
image = Image.open('./assets/radio.png').convert('RGB')
pixel_values = image_processor(images=image, return_tensors='pt', do_resize=True).pixel_values
pixel_values = pixel_values.cuda()
summary, features = model(pixel_values)
```
### Usage
RADIO will return a tuple with two tensors. The `summary` is similar to the `cls_token` in ViT and is meant to represent the general concept of the entire image. It has shape $(B,C)$ with $B$ being the batch dimension, and $C$ being some number of channels. The `spatial_features` represent more localized content which should be suitable for dense tasks such as semantic segmentation, or for integration into an LLM. It has shape $(B,T,D)$ with $T$ being the flattened spatial tokens, and $D$ being the channels for spatial features. Note that $C \neq D$ in general.
Converting to a spatial tensor format can be done using the downsampling size of the model, combined with the input tensor shape. For 'radio_v1', the patch size is 14.
```Python
from einops import rearrange
spatial_features = rearrange(spatial_features, 'b (h w) d -> b d h w', h=x.shape[-2] // patch_size, w=x.shape[-1] // patch_size)
```
The resulting tensor will have shape $(B,D,H,W)$, as is typically seen with computer vision models.
### RADIOv2.5 Notes
See the [RADIOv2.5 technical report](https://github.com/NVlabs/RADIO/blob/main/RADIOv2.5_tech_report.md).
## License
RADIO code and weights are released under the [NSCLv1 License](LICENSE).
## Citing RADIO
If you find this repository useful, please consider giving a star and citation:
```
@InProceedings{Ranzinger_2024_CVPR,
author = {Ranzinger, Mike and Heinrich, Greg and Kautz, Jan and Molchanov, Pavlo},
title = {AM-RADIO: Agglomerative Vision Foundation Model Reduce All Domains Into One},
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2024},
pages = {12490-12500}
}
```
```
@misc{ranzinger2024phisdistributionbalancinglabelfree,
title={PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation},
author={Mike Ranzinger and Jon Barker and Greg Heinrich and Pavlo Molchanov and Bryan Catanzaro and Andrew Tao},
year={2024},
eprint={2410.01680},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2410.01680},
}
```
|