Diffusion model morphing aligns images without tuning or ghosting
AlignMorph: Tuning-Free Diffusion Image Morphing via Explicit Semantic Transport
Computer Vision and Pattern Recognition
Summary
Image morphing smoothly transforms one picture into another, but existing methods struggle when the images differ a lot in layout or require long setup times. The authors propose AlignMorph, which separates the task of matching parts of the images from creating the smooth transition, making the process easier and faster. AlignMorph uses a special way to align image content and then ensures the transformation stays consistent and clear without blurring or odd artifacts. This approach works well on tests without needing extra adjustments.
What this means in practice
- •For graphic designers: Create smooth and clear image transitions for video effects without manual parameter tuning or visual glitches.
- •For game developers: Generate realistic morphing animations between character states or objects with better structural consistency and no ghost images.
Authors
Wuyi Liu, Xu Han, Yuren Chen, Yige Mao, Zishuo Peng, Xianzhi Li
Abstract
Image morphing aims to produce a smooth and semantically consistent transition between two input images. Existing diffusion-based morphing methods either require expensive per-pair optimization or rely on implicit spatial alignment, which easily fails under large layout discrepancies. To address these limitations, we propose AlignMorph, a novel tuning-free diffusion framework guided by the principle of transport-then-denoise. We explicitly decouple geometric alignment from generative denoising to avoid structural entanglement. Our framework consists of two core components. (1) Global Semantic Transport, which achieves diffusion-compatible semantic alignment via entropic optimal transport and reliability-aware latent warping; and (2) Coordinate-Aligned Generation, which uses a symmetric bi-phase attention handoff to maintain consistent spatial coordinates throughout denoising. Without any tuning, AlignMorph effectively eliminates ghosting and achieves superior structural coherence and temporal smoothness on morphing benchmarks. Code is available at https://github.com/51xOne/Alignmorph.