学術雑誌論文

学術雑誌論文

text

Pattern Recognition. 2025, 172, p. 112614

Instance-wise distribution control of text-to-image diffusion models

Elsevier Ltd

英語

text

2025-11-01

https://doi.org/10.1016/j.patcog.2025.112614

Ian Chan W., Santo H., Matsushita Y., et al. Instance-wise distribution control of text-to-image diffusion models. Pattern Recognition 172, Part C, 112614 (2026); https://doi.org/10.1016/j.patcog.2025.112614.

Text-to-image diffusion models are increasingly used to generate synthetic datasets for downstream vision tasks. However, they often inherit biases from large-scale training data, which can result in unbalanced attribute distributions in the generated images. While prior efforts have attempted to mitigate these biases, most focus on single-object images and struggle to control attributes across object instances in multi-instance generations. To address this limitation, we propose an instance-wise control of the attribute distribution by fine-tuning diffusion models with guidance from a pre-trained object detector and an attribute classifier. Our approach aligns the attribute distribution over object instances in generated images with a user-defined distribution, which enables precise control over attribute proportions at the instance level. Experiments across various objects and attributes demonstrate that our method generates high-quality, multi-instance images that match the specified distribution, supporting the scalable creation of distribution-aware synthetic datasets for in-the-wild vision tasks.

This article is licensed under a Creative Commons Attribution 4.0 International License.

103517

Instance-wise distribution control of text-to-image diffusion models

Elsevier Ltd

英語

2025-11-01

Pattern Recognition

172

112614

https://doi.org/10.1016/j.patcog.2025.112614

https://hdl.handle.net/11094/103517