DeeperCut: A Deeper, Stronger, and Faster Multi-Person Pose Estimation Model
Eldar Insafutdinov1 Leonid Pishchulin1 Bjoern Andres1 Mykhaylo Andriluka1,2 and Bernt Schiele1
Abstract
The goal of this paper is to advance the state-of-the-art of articulated pose estimation in scenes with multiple people. To that end we contribute on three fronts. We propose (1) improved body part detectors that generate effective bottom-up proposals for body parts; (2) novel image-conditioned pairwise terms that allow to assemble the proposals into a variable number of consistent body part configurations; and (3) an incremental optimization strategy that explores the search space more efficiently thus leading both to better performance and significant speed-up factors. Evaluation is done on two single-person and two multi-person pose estimation benchmarks. The proposed approach significantly outperforms best known multi-person pose estimation results while demonstrating competitive performance on the task of single person pose estimation111Models and code available at http://pose.mpi-inf.mpg.de.
中文速览
多人场景下的人体姿态估计（articulated pose estimation）一直面临身体部位检测不准、多人之间相互遮挡难以区分、以及推断速度极慢三大难题。这篇论文在已有的DeepCut框架基础上，从三个方向同时发力：用152层残差网络（ResNet）构建更强的身体部位检测器，设计以图像内容为条件的全新成对关系项（image-conditioned pairwise terms）来更准确地判断两个候选部位是否属于同一个人，并提出增量式优化策略来大幅缩短整数线性规划的求解时间。实验结果显示，新方法在多人姿态估计基准上大幅超越此前最优水平，推断速度相比原始DeepCut加速了2至3个数量级，同时在单人姿态估计任务上也保持了极具竞争力的精度。这项工作的意义在于，它让在真实拥挤场景中同时、准确、快速地估计所有人的姿态成为可能，为后续多人姿态估计研究奠定了重要基础。
原文 arXiv:1605.03170;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1605.03170v3