Beyond Rigid: Benchmarking Non-Rigid Video Editing
This work addresses the problem of physics-aware video editing for researchers and practitioners, providing a standard testing platform, though it is incremental as it builds on existing video editing methods.
The paper tackles the challenge of generating coherent non-rigid deformations in text-driven video editing, which often suffers from physical distortion and temporal flicker, by proposing NRVBench, a comprehensive benchmark that includes a dataset of 180 videos, a novel evaluation metric (NRVE-Acc), and a training-free baseline method (VM-Edit) that achieves excellent performance across metrics.
Despite the remarkable progress in text-driven video editing, generating coherent non-rigid deformations remains a critical challenge, often plagued by physical distortion and temporal flicker. To bridge this gap, we propose NRVBench, the first dedicated and comprehensive benchmark designed to evaluate non-rigid video editing. First, we curate a high-quality dataset consisting of 180 non-rigid motion videos from six physics-based categories, equipped with 2,340 fine-grained task instructions and 360 multiple-choice questions. Second, we propose NRVE-Acc, a novel evaluation metric based on Vision-Language Models that can rigorously assess physical compliance, temporal consistency, and instruction alignment, overcoming the limitations of general metrics in capturing complex dynamics. Third, we introduce a training-free baseline, VM-Edit, which utilizes a dual-region denoising mechanism to achieve structure-aware control, balancing structural preservation and dynamic deformation. Extensive experiments demonstrate that while current methods have shortcomings in maintaining physical plausibility, our method achieves excellent performance across both standard and proposed metrics. We believe the benchmark could serve as a standard testing platform for advancing physics-aware video editing.