Generative model uses geometry to keep shapes consistent in images and videos
Geodesic-informed Generative Diffusion Model For Topology-preserved Image Video Generation
Machine LearningComputer Vision and Pattern RecognitionGraphics
Summary
Current image-generating AI models are good at making pictures but often fail to keep the shapes and structures of objects consistent, which can be a big problem for fields like medicine and robotics. The authors created a new system called IGG that focuses on how shapes change smoothly and preserves their structure while generating images and videos. Instead of just working with colors and textures, their model learns to make these shape changes using mathematical paths called geodesics. This approach helps their model create images where the underlying object forms stay reliable and easier to understand.
Generative diffusion modelsImage synthesisTopology preservationGeodesic deformationComputational anatomyShape consistencyInvertible mappingsDeformation dynamicsImage generationSmooth mappings
Authors
Nian Wu, Nivetha Jayakumar, Jiarui Xing, Miaomiao Zhang
Abstract
Generative diffusion models have emerged as a class of powerful techniques for various imaging applications, including but not limited to synthesis, reconstruction, and segmentation. Despite their success, current generative models pose two key limitations. First, they primarily rely on image intensity and texture information, with limited attention to underlying object geometry. As a result, they do not guarantee geometric or topological consistency during the generation process, which is a crucial requirement for high-stakes domains such as computational anatomy, biology, and robotics, where preserving object structure is critical. Second, existing models fail to explicitly learn or represent shape changes in the generative process. Such deformation dynamics remain occluded within network parameters; hence leaving the transformation process uninterpretable and physically uninformed. To address these challenges, we introduce IGG (Image Generation informed by Geodesic dynamics), a novel framework that integrates topology-preserving geodesic principles into the diffusion-based generative process. In contrast to conventional methods that operate in image intensity space, IGG learns and synthesizes diverse samples within geodesic deformation spaces, where geometric object changes are learned as smooth and invertible smooth mappings from a given template/source image. Our code is publicly available at https://github.com/nellie689/IGG.