analiza aiqbench
SciFusion is a research preprint proposing a latent diffusion model that conditions on domain, object, and contrast factors, plus physics-based consistency losses for MRI, CT, and electron microscopy. The reported gains are over modality-specific and flattened baselines across 15 imaging modalities, with zero-shot synthesis of unseen combinations. Readers should check how the physics terms are implemented and whether the zero-shot claims hold up on truly held-out compositions and downstream clinical or scientific tasks.
SciFusion: Physics-Guided Compositional Diffusion for Multimodal 3D Scientific Image Generation
arXiv:2610.30313v1 Announce Type: cross Abstract: Existing 3D generative approaches are typically developed for individual imaging modalities, requiring substantial modality-specific data and limiting their ability to exploit shared structure across scientific imaging domains. We present SciFusion, a multimodal framework for unified 3D scientific image generation built on a latent diffusion backbone, with two key contributions. First, we introduce a compositional conditioning scheme that factorizes each modality into domain, object, and contrast representations. Sharing these factors across related modalities enables knowledge transfer across heterogeneous imaging domains and allows the factors to be recombined at inference time to synthesize unseen domain-object-contrast compositions in a zero-shot manner. Second, we incorporate domain-specific physics-informed objectives: Fourier spectral consistency for MRI, Radon projection consistency for CT, and multislice diffraction consistency for electron microscopy, to encourage generated samples to respect modality-dependent image formation processes. Extensive experiments across 15 medical and scientific imaging modalities demonstrate that SciFusion improves generation fidelity and cross-domain generalization over modality-specific and flattened multimodal conditioning strategies, while enabling zero-shot synthesis of unseen compositions without training on those target compositions. These results demonstrate that compositional, physics-guided conditioning provides a scalable framework for unified 3D scientific image generation, particularly for emerging modalities where large-scale training data are difficult to obtain.
Abstract:Existing 3D generative approaches are typically developed for individual imaging modalities, requiring substantial modality-specific data and limiting their ability to exploit shared structure across scientific imaging domains. We present SciFusion, a multimodal framework for unified 3D scientific image generation built on a latent diffusion backbone, with two key contributions. First, we introduce a compositional conditioning scheme that factorizes each modality into domain, object, and contrast representations. Sharing these factors across related modalities enables knowledge transfer across heterogeneous imaging domains and allows the factors to be recombined at inference time to synthesize unseen domain-object-contrast compositions in a zero-shot manner. Second, we incorporate domain-specific physics-informed objectives: Fourier spectral consistency for MRI, Radon projection consistency for CT, and multislice diffraction consistency for electron microscopy, to encourage generated samples to respect modality-dependent image formation processes. Extensive experiments across 15 medical and scientific imaging modalities demonstrate that SciFusion improves generation fidelity and cross-domain generalization over modality-specific and flattened multimodal conditioning strategies, while enabling zero-shot synthesis of unseen compositions without training on those target compositions. These results demonstrate that compositional, physics-guided conditioning provides a scalable framework for unified 3D scientific image generation, particularly for emerging modalities where large-scale training data are difficult to obtain.
Image and Video Processing (eess.IV); Machine Learning (cs.LG)
arXiv:2610.30313 [eess.IV]
(or arXiv:2610.30313v1 [eess.IV] for this version)
https://doi.org/10.48550/arXiv.2610.30313 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
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.
Text integral sub licența deschisă a publicației · 297 cuvinte