1 paper
David C. Jeong, Hongji Liu, Saunder Salazar +2
While recent two-stage many-to-one deep learning models have demonstrated great success in 3D human pose estimation, such models are inefficient ways to detect 3D key points in a s…