Image Segmentation
Transformers
Safetensors
remote-sensing
earth-observation
open-vocabulary
clip
sam3
semantic-segmentation
Instructions to use Dingyi111/SegEarth-OV with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Dingyi111/SegEarth-OV with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-segmentation", model="Dingyi111/SegEarth-OV")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("Dingyi111/SegEarth-OV", dtype="auto") - Notebooks
- Google Colab
- Kaggle
File size: 1,072 Bytes
fabc606 | 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 | {
"description": "Unified SegEarth: OV, OV-2 (CLIP-based), OV-3 (SAM3-based). Each variant in self-contained subfolder.",
"subfolders": {
"OV": {
"path": "OV",
"config": "OV/config.json",
"backbone": "CLIP",
"model_id": "openai/clip-vit-base-patch16",
"featup": "jbu_one"
},
"OV-2": {
"path": "OV-2",
"config": "OV-2/config.json",
"backbone": "AlignEarth",
"model_id": "BiliSakura/AlignEarth-SAR-ViT-B-16",
"featup": "jbu_one"
},
"OV-3": {
"path": "OV-3",
"config": "OV-3/config.json",
"backbone": "SAM3",
"model_id": "facebook/sam3",
"featup": null
}
},
"usage": {
"OV": "from OV.pipeline import SegEarthPipeline; pipe = SegEarthPipeline()",
"OV-2": "from OV-2.pipeline import SegEarthPipeline; pipe = SegEarthPipeline()",
"OV-3": "from OV-3.pipeline import SegEarthPipeline; pipe = SegEarthPipeline()"
},
"checkpoints": "Self-contained: all backbones (CLIP, AlignEarth, SAM3) and SimFeatUp weights included; no download required"
}
|