ICCV Daily 2021 - Wednesday

The goal of this work is to reconstruct 3D human textures from a single input image . Its novelty is a new framework based on the transformer architecture , which sets it apart from previous works which are based mostly on convolutional neural networks. Xiangyu believes this is the first work to use transformers for 3D human 4 DAILY ICCV Wednesday Oral Presentation 3D Human Texture Estimation from a Single Image with Transformers Xiangyu Xu is a research fellow at Nanyang Technological University, having graduated from Tsinghua University in China. His work, which introduces a novel architecture for reconstructing 3D human textures from a single input image, has been accepted as an oral presentation. He speaks to us ahead of his live Q&A session today.