Human Pose Regression by Combining Indirect Part Detection and Contextual Information

By Diogo Carbonera Luvizon, Hedi Tabia, David Picard. 2017

268 citations recorded by the source corpus.

In this paper, we propose an end-to-end trainable regression approach for human pose estimation from still images. We use the proposed Soft-argmax function to convert feature maps directly to joint coordinates, resulting in a fully differentiable framework. Our method is able to learn heat maps representations indirectly, without additional steps of artificial ground truth generation. Consequently, contextual information can be included to the pose predictions in a seamless way. We evaluated our method on two very challenging datasets, the Leeds Sports Poses (LSP) and the MPII Human Pose datasets, reaching the best performance among all the existing regression methods and comparable results to the state-of-the-art detection based approaches.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

2D/3D Pose Estimation and Action Recognition Using Multitask Deep Learning (2018, methodology): “An intuitive graphical explanation of the Soft-argmax layer is shown in Figure 3.”
Deep High-Resolution Representation Learning for Human Pose Estimation (2019, neutral): “:4 83:6 90:9 Sun et al. [58] 98:1 96:2 91:2 87:2 89:8 87:4 84:1 91:0 Tang et al. [63] 97 :4 96 4 92 1 87 7 90 2 87 7 84 3 91 2 Ning et al. [44] 98 :1 96 3 92 2 87 8 90 6 87 6 82 7 91 2 Luvizon et al. [37] 98:1 96:6 92:0 87:5 90:6 88:0 82:7 91:2 Chu et al. [14] 98:5 96:3 91:9 88:1 90:6 88:0 85:0 91:5”
Integral Human Pose Regression (2017, background): “Speciﬁcally, two contemporary works [29,34] also apply it for human pose estimation.”
Multi-Scale Structure-Aware Network for Human Pose Estimation (2018, unknown): “8.1 96.5 92.5 88.5 90.2 89.6 86.0 91.9 61.6 Chou et al. arXiv’17 [9] 98.2 96.8 92.2 88.0 91.3 89.1 84.9 91.8 63.9 Chu CVPR’17 [12] 98.5 96.3 91.9 88.1 90.6 88.0 85.0 91.5 63.8 Luvizon et al. arXiv’17 [17] 98.1 96.6 92.0 87.5 90.6 88.0 82.7 91.2 63.9 Ning et al. TMM’17 [19] 98.1 96.3 92.2 87.8 90.6 8”
Human Pose Regression with Residual Log-likelihood Estimation (2021, methodology): “These methods generate a likelihood heatmap for each joint and locate the joint as the point with the argmax [56, 64, 46] or soft-argmax [40, 31, 54] operations.”
Deep Learning-based Human Pose Estimation: A Survey (2020, methodology): “, [36] [41] [42] [43] [44] [45] [46] [47] [48] [49]) to predict joint coordinates from images as shown in Fig.”
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].”
Deep 3D human pose estimation: A review (2021, unknown): “It is elucidated by Mehta et al. (2018) that introducing the novel occlusionrobust pose-maps (ORPM) can outputs a fixed number of maps to encode the 3D poses of everyone in the picture.”
Monocular human pose estimation: A survey of deep learning-based methods (2020, background): “…Direct prediction : (Krizhevsky et al., 2012), on video (Pﬁster et al., 2014) (2) Supervision improvement : transform heatmaps to joint coordinates (Luvizon et al., 2017; Nibali et al., 2018), recursive reﬁnement (Carreira et al., 2016), bone-based constraint (Sun et al., 2017) (3) Multi-task :…”