最近，视频生成模型取得了显著进展。与图像生成不同，视频生成面临更大的挑战，不仅需要生成高质量的帧，还必须确保这些帧之间的时间一致性。尽管取得了令人瞩目的进展，但关于评估生成视频质量的指标研究，尤其是在时间和运动一致性方面，仍然探索不足。为了填补这一研究空白，我们提出了 Fréchet Video Motion Distance (FVMD) 指标，专注于评估视频生成中的运动一致性。具体而言，我们基于关键点跟踪设计了显式的运动特征，然后通过 Fréchet 距离来衡量这些特征之间的相似性。我们通过向真实视频注入噪声来进行敏感性分析，以验证 FVMD 的有效性。此外，我们进行了一项大规模的人类研究，表明我们的指标能有效检测时间噪声，并且比现有指标更符合人类对生成视频质量的感知。此外，我们的运动特征能够持续提升 Video Quality Assessment (VQA) 模型的性能，表明我们的方法也适用于一元视频质量评估。代码可在 https://github.com/ljh0v0/FMD-frechet-motion-distance 获取。
英文摘要：
Significant advancements have been made in video generative models recently. Unlike image generation, video generation presents greater challenges, requiring not only generating high-quality frames but also ensuring temporal consistency across these frames. Despite the impressive progress, research on metrics for evaluating the quality of generated videos, especially concerning temporal and motion consistency, remains underexplored. To bridge this research gap, we propose Fréchet Video Motion Distance (FVMD) metric, which focuses on evaluating motion consistency in video generation. Specifically, we design explicit motion features based on key point tracking, and then measure the similarity between these features via the Fréchet distance. We conduct sensitivity analysis by injecting noise into real videos to verify the effectiveness of FVMD. Further, we carry out a large-scale human study, demonstrating that our metric effectively detects temporal noise and aligns better with human perceptions of generated video quality than existing metrics. Additionally, our motion features can consistently improve the performance of Video Quality Assessment (VQA) models, indicating that our approach is also applicable to unary video quality evaluation. Code is available at https://github.com/ljh0v0/FMD-frechet-motion-distance.
TIT-Score: Evaluating Long-Prompt Based Text-to-Image Alignment via Text-to-Image-to-Text Consistency
Institute of Image Communication and Network Engineering（图像通信与网络工程研究所）；MoE Key Lab of Artificial Intelligence, AI Institute（人工智能关键实验室）；Shanghai Jiao Tong University（上海交通大学）
SC3-Eval: Evaluating Robot Foundation Models via Self-Consistent Video Generation
University of Toronto（多伦多大学）；Vector Institute（向量研究所）；NVIDIA（英伟达）；Physical Intelligence；Stanford University（斯坦福大学）；UC Berkeley（加州大学伯克利分校）；Allen Institute for AI（艾伦人工智能研究所）
MIND: Benchmarking Memory Consistency and Action Control in World Models
CSU-JPG, Central South University（中南大学）；National University of Singapore（新加坡国立大学）；Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））；Nanyang Technological University（南洋理工大学）
GUI-GenBench: Evaluating Image Generation Models as Interactive GUI Environments
StepFun；South China University of Technology（南方科技大学）；Peking University（北京大学）；Tsinghua University（清华大学）；Institute of Automation Chinese Academy of Sciences（中国科学院自动化研究所）；The University of Chicago（芝加哥大学）；Nanyang Technological University（南洋理工大学）
Performance vs Consistency: Evaluating a Foundation Model in Lung-RADS Screening
Median Technologies；The University of Osaka（大阪大学）；Université de Paris Cité（巴黎西岱大学）；AP-HP（巴黎公立医院集团）；Hôpital Universitaire Necker Enfants Malades（内克尔儿童医院）；Memorial Sloan Kettering Cancer Center（纪念斯隆凯特琳癌症中心）；Massachusetts General Hospital（马萨诸塞总医院）；Sorbonne Université（索邦大学）
Evaluating Vision-Language Models for Image Quality Assessment using Psychophysical Data
School of Engineering, University of Galway（Galway大学工程学院）；State Key Laboratory of Extreme Photonics and Instrumentation, Zhejiang University（浙江大学极端光信息获取国家重点实验室）
