Zhang Yang, Zhang Hu
Beijing AIQI Technology Co., LTD., Beijing, China.
PeerJ Comput Sci. 2024 Jul 15;10:e2187. doi: 10.7717/peerj-cs.2187. eCollection 2024.
The popularization of intelligent toys enriches the lives of the general public. To provide the public with a better toy experience, we propose the intelligent toy tracking method by the mobile cloud terminal deployment and depth-first search algorithm. Firstly, we construct a toy detection model Transformer, which realizes the positioning of toys in the image through the refined region adaptive boundary representation. Then, using these detected continuous frames, we improve the toy tracking based on a depth-first search. Long-short-term memory (LSTM) constructs the continuous frame tracking structure, and the depth-first search mechanism is embedded to realize the accurate tracking of multiple targets in continuous frames. Finally, to realize the terminal marginalization of the proposed method, this chapter proposes a lightweight model deployment method based on mobile cloud terminals to realize the maintenance of the optimal machine state of intelligent toys. The experiment proves that our proposed target method can reach the world-leading level and obtain the mAP value of 0.858. Our tracking method can also perform excellently with a MOTA value of 0.916.
智能玩具的普及丰富了普通大众的生活。为了给公众提供更好的玩具体验，我们提出了通过移动云终端部署和深度优先搜索算法的智能玩具跟踪方法。首先，我们构建了一个玩具检测模型Transformer，它通过精细区域自适应边界表示实现图像中玩具的定位。然后，利用这些检测到的连续帧，我们基于深度优先搜索改进玩具跟踪。长短期记忆（LSTM）构建连续帧跟踪结构，并嵌入深度优先搜索机制以实现连续帧中多个目标的精确跟踪。最后，为了实现所提方法的终端边缘化，本章提出了一种基于移动云终端的轻量级模型部署方法，以实现智能玩具最优机器状态的维护。实验证明，我们提出的目标方法可以达到世界领先水平，获得0.858的平均精度均值（mAP）值。我们的跟踪方法在多目标跟踪精度（MOTA）值为0.916时也能表现出色。
