MorphAny3D: Unleashing the Power of Structured Latent in 3D Morphing

2026-01-01

arXiv.org

AI Summary

This work proposes a training-free deformation framework to address semantic inconsistency and temporal discontinuity in cross-category 3D shape morphing. By operating in the Structured Latent Attribute Space (SLAT), the method introduces Morphing Cross-Attention to effectively fuse source and target features, while integrating Temporal-Fused Self-Attention and a pose correction strategy to ensure temporally smooth and structurally coherent deformations. The framework enables disentangled shape deformation and 3D style transfer, and generalizes seamlessly to other SLAT-based models. Extensive experiments demonstrate that it achieves state-of-the-art results across diverse cross-category scenarios, significantly improving both semantic consistency and temporal smoothness in the generated 3D morphing sequences.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Lexical Semantics and Morphology
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

3D morphing remains challenging due to the difficulty of generating semantically consistent and temporally smooth deformations, especially across categories. We present MorphAny3D, a training-free framework that leverages Structured Latent (SLAT) representations for high-quality 3D morphing. Our key insight is that intelligently blending source and target SLAT features within the attention mechanisms of 3D generators naturally produces plausible morphing sequences. To this end, we introduce Morphing Cross-Attention (MCA), which fuses source and target information for structural coherence, and Temporal-Fused Self-Attention (TFSA), which enhances temporal consistency by incorporating features from preceding frames. An orientation correction strategy further mitigates the pose ambiguity within the morphing steps. Extensive experiments show that our method generates state-of-the-art morphing sequences, even for challenging cross-category cases. MorphAny3D further supports advanced applications such as decoupled morphing and 3D style transfer, and can be generalized to other SLAT-based generative models. Project page: https://xiaokunsun.github.io/MorphAny3D.github.io/.

Problem

• 3D morphing
• semantic consistency
• temporal smoothness
• cross-category deformation

Innovation

• Structured Latent
• 3D Morphing
• Cross-Attention
• Temporal Consistency
• Training-Free Framework