ABSTRACT
Realistic animation of tree branches under wind remains challenging. This paper introduces Temporal Consistency Fusion (TCF), a neural physics framework that integrates graph-based spatial encoding with temporal attention mechanisms for efficient and coherent branch motion synthesis.
PAPER · PDF
Loading PDF...
Key findings
TCF achieves 3.2x speedup over traditional physics engines.
Reduces temporal inconsistency metrics by 67% compared to neural approaches.
Supports real-time performance for trees with 1,000+ branches.
Limitations & open questions
The framework's scalability to larger environments and more complex interactions is not discussed.