FaceInstruct-1M
收藏arXiv2025-09-30 收录
下载链接：
https://face-llava.github.io
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个以人脸为中心的数据库，旨在对多模态大型语言模型（MLLMs）进行指令调优，以处理人脸相关任务。该数据集包含一百万个样本，被用于构建在多个基准测试和任务上的性能评估，包括面部表情和属性识别、动作单元检测、年龄估计以及深度伪造检测等任务。
This dataset is a face-centric database intended for instruction tuning of multimodal large language models (MLLMs) to handle face-related tasks. It contains one million samples and is utilized for performance evaluation across multiple benchmarks and tasks, including facial expression and attribute recognition, action unit detection, age estimation, and deepfake detection.
