hexuan21/VideoFeedback_bad5k

该数据集包含两个配置：annotated和real。每个配置包含以下特征：id（字符串类型）、images（字符串序列）、text prompt（字符串类型）、video link（字符串类型）、visual quality（int64类型）、temporal consistency（int64类型）、dynamic degree（int64类型）、text-to-video alignment（int64类型）、factual consistency（int64类型）以及conversations（包含from和value两个字段的列表）。数据集主要用于评估视频生成模型的质量，包括视觉质量、时间一致性、动态程度、文本到视频的对齐度以及事实一致性等方面。

The dataset includes two configurations: annotated and real. Each configuration contains the following features: id (string type), images (sequence of strings), text prompt (string type), video link (string type), visual quality (int64 type), temporal consistency (int64 type), dynamic degree (int64 type), text-to-video alignment (int64 type), factual consistency (int64 type), and conversations (a list containing fields from and value). The dataset is primarily used for evaluating the quality of video generation models, including aspects such as visual quality, temporal consistency, dynamic degree, text-to-video alignment, and factual consistency.