ReMatching Dynamic Reconstruction Flow
This work addresses a fundamental computer vision task for applications requiring high-quality dynamic scene reconstruction, though it appears incremental as it augments existing methods.
The paper tackles the problem of reconstructing dynamic scenes from images by introducing the ReMatching framework, which incorporates deformation priors into existing models, resulting in clear improvements in reconstruction accuracy on benchmarks.
Reconstructing a dynamic scene from image inputs is a fundamental computer vision task with many downstream applications. Despite recent advancements, existing approaches still struggle to achieve high-quality reconstructions from unseen viewpoints and timestamps. This work introduces the ReMatching framework, designed to improve reconstruction quality by incorporating deformation priors into dynamic reconstruction models. Our approach advocates for velocity-field based priors, for which we suggest a matching procedure that can seamlessly supplement existing dynamic reconstruction pipelines. The framework is highly adaptable and can be applied to various dynamic representations. Moreover, it supports integrating multiple types of model priors and enables combining simpler ones to create more complex classes. Our evaluations on popular benchmarks involving both synthetic and real-world dynamic scenes demonstrate that augmenting current state-of-the-art methods with our approach leads to a clear improvement in reconstruction accuracy.