Decoupling High and Low Frequencies for Faithful Image Generation with Fine Details

Abstract:Latent generative models compress images into learned embeddings prior to synthesis, and the generation quality critically depends on how faithfully these embeddings preserve visual detail. We observe that while such embeddings are effective at reconstructing low frequency structure, they struggle to recover sharp high frequency details that are essential for perceptual realism. Conventional reconstruction objectives implicitly prioritize coarse structural information over high frequency content, which can lead to overly smoothed outputs and degraded visual quality in textured regions. Motivated by this observation, we propose DeBaT, a Decoupled frequency Band Tokenizer that explicitly separates the learning of low and high frequency band embeddings. This decoupling enables accurate reconstruction of fine details while preserving global coherence. Integrated into a latent diffusion based generative model, DeBaT allows for sharper and more realistic samples than previous latent tokenizers, confirming that the explicit decoupling of high and low frequency bands eases the preservation of visual details in learned embedding spaces.

Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)

arXiv:2509.05441 [cs.CV]

(or arXiv:2509.05441v4 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2509.05441 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.