Mask Generation
Transformers
Safetensors
sam2
sam2_video
feature-extraction
libreyolo
promptable-segmentation
image-segmentation
Instructions to use LibreYOLO/LibreSAM2tiny with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use LibreYOLO/LibreSAM2tiny with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("mask-generation", model="LibreYOLO/LibreSAM2tiny")# Load model directly from transformers import AutoProcessor, AutoModel processor = AutoProcessor.from_pretrained("LibreYOLO/LibreSAM2tiny") model = AutoModel.from_pretrained("LibreYOLO/LibreSAM2tiny") - sam2
How to use LibreYOLO/LibreSAM2tiny with sam2:
# Use SAM2 with images import torch from sam2.sam2_image_predictor import SAM2ImagePredictor predictor = SAM2ImagePredictor.from_pretrained(LibreYOLO/LibreSAM2tiny) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): predictor.set_image(<your_image>) masks, _, _ = predictor.predict(<input_prompts>)# Use SAM2 with videos import torch from sam2.sam2_video_predictor import SAM2VideoPredictor predictor = SAM2VideoPredictor.from_pretrained(LibreYOLO/LibreSAM2tiny) with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16): state = predictor.init_state(<your_video>) # add new prompts and instantly get the output on the same frame frame_idx, object_ids, masks = predictor.add_new_points(state, <your_prompts>): # propagate the prompts to get masklets throughout the video for frame_idx, object_ids, masks in predictor.propagate_in_video(state): ... - Notebooks
- Google Colab
- Kaggle
| license: apache-2.0 | |
| library_name: transformers | |
| pipeline_tag: mask-generation | |
| tags: | |
| - libreyolo | |
| - sam2 | |
| - promptable-segmentation | |
| - image-segmentation | |
| base_model: facebook/sam2.1-hiera-tiny | |
| # LibreSAM2tiny | |
| SAM-2.1 Hiera Tiny rehosted for LibreYOLO's `LibreSAM` promptable segmentation tier. | |
| ## Source | |
| Derived from [`facebook/sam2.1-hiera-tiny`](https://huggingface.co/facebook/sam2.1-hiera-tiny) at commit | |
| `de431c4043854a71d8101e17995dfe596bf101a5` and the Apache-2.0 | |
| [`facebookresearch/sam2`](https://github.com/facebookresearch/sam2) source | |
| release. | |
| ## Modifications | |
| Learned parameters are unchanged. The upstream Transformers-compatible snapshot | |
| files are mirrored here for LibreYOLO distribution. This repository adds | |
| LibreYOLO model-card packaging plus `LICENSE` and `NOTICE` files for Apache-2.0 | |
| redistribution. | |
| ## Usage | |
| ```python | |
| from libreyolo import LibreSAM | |
| model = LibreSAM("sam2-tiny") | |
| result = model("image.jpg", points=[500, 375], labels=[1]) | |
| ``` | |
| ## License | |
| Apache License 2.0. See [`LICENSE`](./LICENSE) and [`NOTICE`](./NOTICE). | |