HRPose: Real-Time High-Resolution 6D Pose Estimation Network Using Knowledge Distillation

作     者：GUAN Qi SHENG Zihao XUE Shibei GUAN Qi;SHENG Zihao;XUE Shibei

作者机构：Department of Automation Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai Engineering Research Center of Intelligent Control and Management 

出 版 物：《Chinese Journal of Electronics》 (电子学报(英文))

年 卷 期：2023年第32卷第1期

页      面：189-198页

核心收录：

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：supported by the National Natural Science Foundation of China (61873162, 61973317) Open Research Project of the State Key Laboratory of Industrial Control Technology,Zhejiang University,China (ICT2022B47) 

主　　题：Knowledge engineering Computational modeling Pose estimation Grasping Benchmark testing Feature extraction Real-time systems 

摘      要：Real-time six degrees-of-freedom(6D)object pose estimation is essential for many real-world applications, such as robotic grasping and augmented reality. To achieve an accurate object pose estimation from RGB images in real-time, we propose an effective and lightweight model, namely high-resolution 6D pose estimation network(HRPose). We adopt the efficient and small HRNetV2-W18 as a feature extractor to reduce computational burdens while generating accurate 6D poses. With only 33% of the model size and lower computational costs, our HRPose achieves comparable performance compared with state-of-the-art models. Moreover, by transferring knowledge from a large model to our proposed HRPose through output and feature-similarity distillations, the performance of our HRPose is improved in effectiveness and efficiency. Numerical experiments on the widely-used benchmark LINEMOD demonstrate the superiority of our proposed HRPose against state-of-the-art methods.