Intel/Video_Summarization_For_Retail

该数据集名为Video Summarization For Retail Dataset，主要包含零售场景中的购物者短视频及其对应的文本描述。数据集由Parker Lischwe策划，语言为英语，使用cc-by-sa-4.0许可证。数据集的主要用途是微调视觉语言模型，以识别零售场景中的行为和细节。数据集的创建动机是为了改进零售领域的损失预防和盗窃检测系统。需要注意的是，数据集中包含个人可识别信息，且数据集规模不足以可靠地训练模型。

This dataset contains short videos of shoppers in a retail setting along with the corresponding textual descriptions. It is suitable for summarization and visual question answering tasks and is licensed under cc-by-sa-4.0. The dataset supports finetuning of vision language models to recognize behaviors and details in retail scenarios. It includes personally identifiable information in the videos and suggests that the dataset is not comprehensive enough for reliable training of models in understanding visual details within shopping scenarios.