Publications by authors named "Chuanbao He"
Article Synopsis
- This work addresses the issue of low accuracy in recognizing badminton movement trajectories by enhancing the visual system used in badminton robots.
- It employs a convolutional neural network with an attention mechanism to effectively track flying badminton in video streams, while improving the Tiny YOLOv2 detection network to better handle small targets.
- The results show high performance, with an average tracking accuracy of 91.40% and superior metrics in various scenarios compared to traditional algorithms, significantly aiding badminton movement recognition.
