Model description

CoMPaSS-SD1.4

Model Details

• Base Model: StableDiffusion 1.4
• Training Data: SCOP dataset (curated from COCO)
• Framework: Diffusers
• License: Apache-2.0 (see ./LICENSE)

Intended Use

• Generating images with accurate spatial relationships between objects
• Creating compositions that require specific spatial arrangements
• Enhancing the base model's spatial understanding while maintaining its other capabilities

Performance

Key Improvements

• VISOR benchmark: +205.2% relative improvement
• T2I-CompBench Spatial: +183.3% relative improvement
• GenEval Position: +1443.3% relative improvement
• Maintains or improves base model's image fidelity (lower FID and CMMD scores than base model)

Using the Model

Effective Prompting

• Clear spatial relationship descriptors (left, right, above, below)
• Pairs of distinct objects
• Explicit spatial relationships (e.g., "a photo of A to the right of B")

Training Details

Training Data

• Built using the SCOP (Spatial Constraints-Oriented Pairing) data engine
• ~28,000 curated object pairs from COCO
• Enforces criteria for:
  • Visual significance
  • Semantic distinction
  • Spatial clarity
  • Object relationships
  • Visual balance

Training Process

• Trained for 24,000 steps
• Effective batch size of 4
• Learning rate: 5e-6
• Optimizer: AdamW with β₁=0.9, β₂=0.999
• Weight decay: 1e-2

Evaluation Results

Citation

Contact

Download model