D3DFACS

该数据集名为D3DFACS，包含了10位受试者，每位受试者表现出不同数量的面部表情，这些表情都标注了所表现表情激活的动作单元。值得注意的是，这些表情与CoMA数据集中的表情存在显著差异。规模上，该数据集涉及10位受试者，表情数量不等。其任务是进行动态三维面部表情生成。

The dataset is named D3DFACS. It includes 10 subjects, each of whom displays a varying number of facial expressions, and all these expressions are annotated with the action units activated during their performance. Notably, these facial expressions differ significantly from those in the CoMA dataset. In terms of scale, this dataset involves 10 subjects with unequal quantities of facial expressions. The core task targeted by this dataset is dynamic 3D facial expression generation.