<ul class="dashed" data-apple-notes-indent-amount="0"><li><span style="font-family: '.PingFangUITextSC-Regular'">文章标题:</span>15M Multimodal Facial Image-Text Dataset</li><li><span style="font-family: '.PingFangSC-Regular'">文章地址:</span><a href="https://arxiv.org/pdf/2407.08515">https://arxiv.org/abs/2407.08515</a> </li><li>arxiv</li><li>数据集地址:<a href="https://huggingface.co/datasets/OpenFace-CQUPT/FaceCaption-15M">https://huggingface.co/datasets/OpenFace-CQUPT/FaceCaption-15M</a> </li></ul> 概览: <img src="https://res.cloudinary.com/montaigne-io/image/upload/v1728790371/20F68ADD-76A5-4A4C-866B-36A418B2F1D6.png" style="background-color:initial;max-width:min(100%,2978px);max-height:min(1214px);;background-image:url(https://res.cloudinary.com/montaigne-io/image/upload/v1728790371/20F68ADD-76A5-4A4C-866B-36A418B2F1D6.png);height:auto;width:100%;object-fit:cover;background-size:cover;display:block;" width="2978" height="1214"> 与现有数据集比较: <img src="https://res.cloudinary.com/montaigne-io/image/upload/v1728790555/45BE210E-EDE7-4CD6-9ACC-1BF1C76723BF.png" style="background-color:initial;max-width:min(100%,2972px);max-height:min(1020px);;background-image:url(https://res.cloudinary.com/montaigne-io/image/upload/v1728790555/45BE210E-EDE7-4CD6-9ACC-1BF1C76723BF.png);height:auto;width:100%;object-fit:cover;background-size:cover;display:block;" width="2972" height="1020"> 数据集构建方法: <img src="https://res.cloudinary.com/montaigne-io/image/upload/v1728790555/2325789A-FFB3-4695-92B1-AC4854D39AAB.png" style="background-color:initial;max-width:min(100%,2974px);max-height:min(1298px);;background-image:url(https://res.cloudinary.com/montaigne-io/image/upload/v1728790555/2325789A-FFB3-4695-92B1-AC4854D39AAB.png);height:auto;width:100%;object-fit:cover;background-size:cover;display:block;" width="2974" height="1298">