VectorSynth-COSA
is a ControlNet model that generates satellite imagery from OpenStreetMap (OSM) vector data embeddings. It conditions
Stable Diffusion 2.1 Base
on rendered OSM text using the COSA (Contrastive OSM-Satellite Alignment) embedding space.
Model Description
VectorSynth-COSA uses a two-stage pipeline:
RenderEncoder
: Projects 768-dim COSA embeddings to 3-channel control images
ControlNet
: Conditions Stable Diffusion 2.1 on the rendered control images
This model uses COSA embeddings for improved semantic alignment between OSM text and satellite imagery. For the standard CLIP embedding variant, see
VectorSynth
.
Citation
@inproceedings{cher2025vectorsynth,
title={VectorSynth: Fine-Grained Satellite Image Synthesis with Structured Semantics},
author={Cher, Daniel and Wei, Brian and Sastry, Srikumar and Jacobs, Nathan},
year={2025},
eprint={arXiv:2511.07744},
note={arXiv preprint}
}
VectorSynth-COSA huggingface.co is an AI model on huggingface.co that provides VectorSynth-COSA's model effect (), which can be used instantly with this MVRL VectorSynth-COSA model. huggingface.co supports a free trial of the VectorSynth-COSA model, and also provides paid use of the VectorSynth-COSA. Support call VectorSynth-COSA model through api, including Node.js, Python, http.
VectorSynth-COSA huggingface.co is an online trial and call api platform, which integrates VectorSynth-COSA's modeling effects, including api services, and provides a free online trial of VectorSynth-COSA, you can try VectorSynth-COSA online for free by clicking the link below.
MVRL VectorSynth-COSA online free url in huggingface.co:
VectorSynth-COSA is an open source model from GitHub that offers a free installation service, and any user can find VectorSynth-COSA on GitHub to install. At the same time, huggingface.co provides the effect of VectorSynth-COSA install, users can directly use VectorSynth-COSA installed effect in huggingface.co for debugging and trial. It also supports api for free installation.