Establishing a Unified Evaluation Framework for Human Motion Generation: A Comparative Analysis of Metrics
This provides a practical evaluation framework for researchers in human motion generation, though it is incremental as it builds on existing metrics.
The paper tackles the lack of standardized evaluation in human motion generation by reviewing eight metrics, proposing a unified framework, and introducing a novel metric for temporal distortion diversity. It demonstrates this framework by analyzing three generative models on a public dataset.
The development of generative artificial intelligence for human motion generation has expanded rapidly, necessitating a unified evaluation framework. This paper presents a detailed review of eight evaluation metrics for human motion generation, highlighting their unique features and shortcomings. We propose standardized practices through a unified evaluation setup to facilitate consistent model comparisons. Additionally, we introduce a novel metric that assesses diversity in temporal distortion by analyzing warping diversity, thereby enhancing the evaluation of temporal data. We also conduct experimental analyses of three generative models using a publicly available dataset, offering insights into the interpretation of each metric in specific case scenarios. Our goal is to offer a clear, user-friendly evaluation framework for newcomers, complemented by publicly accessible code.