SynLayers Real-World Decomposition

Upload a single image and run the full pipeline in one step:

VLM for whole-caption + bounding-box detection
SynLayers real-image layer decomposition

This Space is intended to run on GPU hardware. The first request may take time while model assets are loaded from Hugging Face.

Input Image

Optional Sample Name

VLM Max New Tokens

128 2048

Seed (-1 keeps config default)

Detected Bounding Boxes

Merged Decomposition

Whole Caption

BBox JSON

Inference Metadata

Predicted Layers

Download Result Bundle

Saved Case Directory