Pose for Action - Action for Pose

By Martin Garbade, Juergen Gall. 2016

103 citations recorded by the source corpus.

In this work we propose to utilize information about human actions to improve pose estimation in monocular videos. To this end, we present a pictorial structure model that exploits high-level information about activities to incorporate higher-order part dependencies by modeling action specific appearance models and pose priors. However, instead of using an additional expensive action recognition framework, the action priors are efficiently estimated by our pose estimation framework. This is achieved by starting with a uniform action prior and updating the action prior during pose estimation. We also show that learning the right amount of appearance sharing among action classes improves the pose estimation. We demonstrate the effectiveness of the proposed method on two challenging datasets for pose estimation and action recognition with over 80,000 test images.

Subfields: 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 Computer Vision and Pattern Recognition

1. Hanqing Lu
2. Yifan Zhang
3. Yu Qiao
4. Yali Wang
5. Junsong Yuan
6. Luc Van Gool
7. Mengyuan Liu
8. Wenbin Du

Citations in the corpus

• 2D/3D Pose Estimation and Action Recognition Using Multitask Deep Learning (2018, neutral): “Pose estimation and action recognition are usually handled as distinct problems [14] or the last is used as a prior for the first [57, 22].”
• Body Joint Guided 3-D Deep Convolutional Descriptors for Action Recognition (2017, methodology): “MST-AOG [40] 0.453 - Pose+DT [34] 0.529 - ST-AOG [41] 0.612 - P-CNN [30] 0.668 0.725 ACPS+IDT-FV [39] 0 DT [42] 0.734 - STIP [35] 0.829 - IDT-FV [11] 0.920 - C3D fc6 [5] 0.860 - Pose features ACPS [39] 0.790 - Actemes [35] 0.794 - Action Bank [43] 0.839 -”
• RPAN: An End-to-End Recurrent Pose-Attention Network for Action Recognition in Videos (2017, methodology): “Alternatively, human poses have proven useful for action recognition [14, 15, 17, 25, 40].”
• Dynamic Kernel Distillation for Efficient Pose Estimation in Videos (2019, unknown): “In [14], Iqbal et al.”
• Recognizing Human Actions as the Evolution of Pose Estimation Maps (2018, methodology): “AOG [48] and Pose + IDT-FV [19] benefit from treating pose estimation and action recognition as a uniform framework, and achieve accuracy of 85.”