arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1901.04877cs.CV
收藏

用于3D姿态估计的特征提升网络

Feature Boosting Network For 3D Pose Estimation

Nanyang Technological University(南洋理工大学);Chalmers University of Technology(查尔姆斯理工大学);Peking University(北京大学);Alibaba Group(阿里巴巴集团)

Jun Liu, Henghui Ding, Amir Shahroudy, Ling-Yu Duan, Xudong Jiang, Gang Wang, Alex C. Kot

摘要:

本文提出了一种特征提升网络,用于从单个RGB图像中估计3D手部姿势和3D身体姿势。在这种方法中,卷积层学习到的特征通过一个新的长短期依赖性感知(LSTD)模块得到提升,该模块使中间卷积特征映射能够使用设计的图形ConvLSTM感知不同手部(或身体)部位之间的图形长短期依赖性。由于3D姿态估计的实际应用中的模糊性、纹理和光照变化以及自遮挡,学习一组可靠且有区别地表示手(或身体)部分的姿态的特征是困难的。为了提高表示每个身体部位的特征的可靠性并增强LSTD模块,本文进一步引入上下文一致性门(CCG),根据其与上下文表示的一致性来调制卷积特征映射。我们评估所提出的方法具有挑战性的基准数据集的3D手部姿势估计和3D全身姿势估计。实验结果表明,我们的方法的有效性,实现国家的最先进的性能上的两个任务。

英文摘要:

In this paper, a feature boosting network is proposed for estimating 3D hand pose and 3D body pose from a single RGB image. In this method, the features learned by the convolutional layers are boosted with a new long short-term dependence-aware (LSTD) module, which enables the intermediate convolutional feature maps to perceive the graphical long short-term dependency among different hand (or body) parts using the designed Graphical ConvLSTM. Learning a set of features that are reliable and discriminatively representative of the pose of a hand (or body) part is difficult due to the ambiguities, texture and illumination variation, and self-occlusion in the real application of 3D pose estimation. To improve the reliability of the features for representing each body part and enhance the LSTD module, we further introduce a context consistency gate (CCG) in this paper, with which the convolutional feature maps are modulated according to their consistency with the context representations. We evaluate the proposed method on challenging benchmark datasets for 3D hand pose estimation and 3D full body pose estimation. Experimental results show the effectiveness of our method that achieves state-of-the-art performance on both of the tasks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑