Ziqi Huang

Hi, we conduct pair-wise comparisons of two sampled videos, using both VBench and Human Preference Annotations. We then compute the spearman correlation of the win ratios on all pairs drawn...

Hi, we're closing this issue as it appears your questions have been addressed. However, feel free to open a new issue or reopen this one, if you have further questions...

Hi, thanks for this question. Please see the model settings here: https://github.com/Vchitect/VBench/tree/master/sampled_videos#model-setting, and these video lengths and frame counts (several seconds) are all supported by VBench. For very long videos...

For models we evaluated, please find their detailed settings here: https://github.com/Vchitect/VBench/tree/master/sampled_videos#what-are-the-details-of-the-video-generation-models

Hi, we're closing this issue as it appears your questions have been addressed. However, feel free to open a new issue or reopen this one, if you have further questions...

Hi, we're closing this issue as it appears your questions have been addressed. However, feel free to open a new issue or reopen this one, if you have further questions...

Hi, thanks for your question! The score range for all dimensions are 0 to 1. For samples of different scores, at different dimensions, you can refer to section G of...

Hi, I assume you are asking about the scores for individual videos in the generated `eval_results.json` file. For `dynamic_degree`, each video undergoes binary classification, with `true` referring to `dynamic`, while...

Hi, it's not probability-based classification, but based on threshold.