Human Pose Regression with Residual Log-likelihood Estimation

By Jiefeng Li, Siyuan Bian, Ailing Zeng, Can Wang, Bo Pang, Wentao Liu, Cewu Lu. IEEE International Conference on Computer Vision, 2021

300 citations recorded by the source corpus.

Heatmap-based methods dominate in the field of human pose estimation by modelling the output distribution through likelihood heatmaps. In contrast, regression-based methods are more efficient but suffer from inferior performance. In this work, we explore maximum likelihood estimation (MLE) to develop an efficient and effective regression-based method. From the perspective of MLE, adopting different regression losses is making different assumptions about the output density function. A density function closer to the true distribution leads to a better regression performance. In light of this, we propose a novel regression paradigm with Residual Log-likelihood Estimation (RLE) to capture the underlying output distribution. Concretely, RLE learns the change of the distribution instead of the unreferenced underlying distribution to facilitate the training process. With the proposed reparameterization design, our method is compatible with off-the-shelf flow models. The proposed method is effective, efficient and flexible. We show its potential in various human pose estimation tasks with comprehensive experiments. Compared to the conventional regression paradigm, regression with RLE bring 12.4 mAP improvement on MSCOCO without any test-time overhead. Moreover, for the first time, especially on multi-person pose estimation, our regression method is superior to the heatmap-based methods. Our code is available at https://github.com/Jeff-sjtu/res-loglikelihood-regression.

View Paper: https://www.semanticscholar.org/paper/ee894195a8135bd4e189bfb77fe0dc9f844c5e2a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Ce Zheng
2. Jiefeng Li
3. Haoshu Fang
4. Tao Jiang
5. Sijie Zhu
6. Wenhan Wu
7. Taojiannan Yang
8. Peng Lu

Citations in the corpus

• RTMPose: Real-Time Multi-Person Pose Estimation based on MMPose (2023, methodology): “Fourth, we revisit the training settings in previous works [3,30,42], and empirically introduce a collection of training strategies applicable to the pose estimation task.”
• Deep Learning-based Human Pose Estimation: A Survey (2020, methodology): “[113] proposed a normalizing low model named RLE (Log-likelihood Estimation) to capture the distribution of joint location, aiming for inding the optimized parameters by residual log-likelihood estimation.”
• AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time (2022, methodology): “Prior works [45], [46] adopt soft-max operation, which works well in single-person pose estimation but remains a large performance gap with the state-of-the-arts in multiperson pose estimation [31], [45], [60].”