Feedback-based object detection for multi-person pose estimation
作者：
Highlights：
• A novel method for increasing the performance through adjusting the object detector based on a new loss function is proposed.
• It is based on an optimization through the estimation results of the keypoint estimator.
• It achieved an accuracy of 74.2 average precision (AP), which is higher than the state-of-the-arts model including the human detector used in the experiment.
• It can perform real-time operations with a high frame rate similar to that of the baseline model.
摘要
•A novel method for increasing the performance through adjusting the object detector based on a new loss function is proposed.•It is based on an optimization through the estimation results of the keypoint estimator.•It achieved an accuracy of 74.2 average precision (AP), which is higher than the state-of-the-arts model including the human detector used in the experiment.•It can perform real-time operations with a high frame rate similar to that of the baseline model.
论文关键词：Pose estimation,Object detection,Multi-person,Real-time system
论文评审过程：Received 23 January 2021, Revised 4 September 2021, Accepted 16 September 2021, Available online 23 September 2021, Version of Record 1 October 2021.
论文官网地址：https://doi.org/10.1016/j.image.2021.116508
