Inject Where It Matters: Training-Free Spatially-Adaptive Identity Preservation for Text-to-Image Personalization
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908835297361920 |
|---|---|
| author | Li, Guandong Ye, Mengxia |
| author_facet | Li, Guandong Ye, Mengxia |
| contents | Personalized text-to-image generation aims to integrate specific identities into arbitrary contexts. However, existing tuning-free methods typically employ Spatially Uniform Visual Injection, causing identity features to contaminate non-facial regions (e.g., backgrounds and lighting) and degrading text adherence. To address this without expensive fine-tuning, we propose SpatialID, a training-free spatially-adaptive identity modulation framework. SpatialID fundamentally decouples identity injection into face-relevant and context-free regions using a Spatial Mask Extractor derived from cross-attention responses. Furthermore, we introduce a Temporal-Spatial Scheduling strategy that dynamically adjusts spatial constraints - transitioning from Gaussian priors to attention-based masks and adaptive relaxation - to align with the diffusion generation dynamics. Extensive experiments on IBench demonstrate that SpatialID achieves state-of-the-art performance in text adherence (CLIP-T: 0.281), visual consistency (CLIP-I: 0.827), and image quality (IQ: 0.523), significantly eliminating background contamination while maintaining robust identity preservation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_13994 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Inject Where It Matters: Training-Free Spatially-Adaptive Identity Preservation for Text-to-Image Personalization Li, Guandong Ye, Mengxia Computer Vision and Pattern Recognition Personalized text-to-image generation aims to integrate specific identities into arbitrary contexts. However, existing tuning-free methods typically employ Spatially Uniform Visual Injection, causing identity features to contaminate non-facial regions (e.g., backgrounds and lighting) and degrading text adherence. To address this without expensive fine-tuning, we propose SpatialID, a training-free spatially-adaptive identity modulation framework. SpatialID fundamentally decouples identity injection into face-relevant and context-free regions using a Spatial Mask Extractor derived from cross-attention responses. Furthermore, we introduce a Temporal-Spatial Scheduling strategy that dynamically adjusts spatial constraints - transitioning from Gaussian priors to attention-based masks and adaptive relaxation - to align with the diffusion generation dynamics. Extensive experiments on IBench demonstrate that SpatialID achieves state-of-the-art performance in text adherence (CLIP-T: 0.281), visual consistency (CLIP-I: 0.827), and image quality (IQ: 0.523), significantly eliminating background contamination while maintaining robust identity preservation. |
| title | Inject Where It Matters: Training-Free Spatially-Adaptive Identity Preservation for Text-to-Image Personalization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.13994 |