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
| { | |
| "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" | |
| } | |