Barycenters of Natural Images - Constrained Wasserstein Barycenters for Image Morphing
Abstract
Image interpolation, or image morphing, refers to a visual transition between two (or more) input images. For such a transition to look visually appealing, its desirable properties are (i) to be smooth; (ii) to apply the minimal required change in the image; and (iii) to seem "real", avoiding unnatural artifacts in each image in the transition. To obtain a smooth and straightforward transition, one may adopt the well-known Wasserstein Barycenter Problem (WBP). While this approach guarantees minimal changes under the Wasserstein metric, the resulting images might seem unnatural. In this work, we propose a novel approach for image morphing that possesses all three desired properties. To this end, we define a constrained variant of the WBP that enforces the intermediate images to satisfy an image prior. We describe an algorithm that solves this problem and demonstrate it using the sparse prior and generative adversarial networks.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
5 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Energy-Guided Continuous Entropic Barycenter Estimation for General Costs cites Wasserstein Iterative Networks for Barycenter Estimation
- Computing Optimal Transport Maps and Wasserstein Barycenters Using Conditional Normalizing Flows cites Wasserstein Iterative Networks for Barycenter Estimation
- Probability Density Geodesics in Image Diffusion Latent Space cites IMPUS: Image Morphing with Perceptually-Uniform Sampling Using Diffusion Models
Cited by top-tier papers5
- Wasserstein Iterative Networks for Barycenter EstimationAlexander Korotin, Vage Egiazarian, Lingxiao Li, Evgeny BurnaevNeurIPS 2022 · 34 citations
- IMPUS: Image Morphing with Perceptually-Uniform Sampling Using Diffusion ModelsZhaoyuan Yang, Zhengyang Yu, Zhiwei Xu, Jaskirat Singh et al.ICLR 2024 · 25 citations
- Energy-Guided Continuous Entropic Barycenter Estimation for General CostsAlexander Kolesov, Petr Mokrov, Igor Udovichenko, Milena Gazdieva et al.NeurIPS 2024 · 13 citations
- Computing Optimal Transport Maps and Wasserstein Barycenters Using Conditional Normalizing FlowsGabriele Visentin, Patrick CheriditoICML 2025
- Probability Density Geodesics in Image Diffusion Latent SpaceQingtao Yu, Jaskirat Singh, Zhaoyuan Yang, Peter Henry Tu et al.CVPR 2025
Related papers
- DiffMorpher: Unleashing the Capability of Diffusion Models for Image MorphingKaiwen Zhang, Yifan Zhou, Xudong Xu, Bo Dai et al.CVPR 2024 · 26 citations
- Textured 3D Regenerative Morphing with 3D Diffusion PriorSonglin Yang, Yushi Lan, Honghua Chen, Xingang PanICCV 2025 · 1 citation
- Interp3D: Correspondence-aware Interpolation for Generative Textured 3D MorphingXiaolu Liu, Yicong Li, Qiyuan He, Jiayin Zhu et al.ICLR 2026 · 6 citations
- Estimating Barycenters of Distributions with Neural Optimal TransportAlexander Kolesov, Petr Mokrov, Igor Udovichenko, Milena Gazdieva et al.ICML 2024 · 13 citations
- Wasserstein-2 Generative NetworksAlexander Korotin, Vage Egiazarian, Arip Asadulaev, Alexander Safin et al.ICLR 2021 · 128 citations
