2D Animal Skeletons Keypoint Detection: Research Progress and Future Trends

文献类型： 外文期刊

第一作者： Ma, Pengfei

作者： Ma, Pengfei;Gao, Ronghua;Huang, Weiwei;Li, Xuwen;Gao, Ronghua;Li, Qifeng;Yu, Qinyang;Wang, Rong;Lai, Chengrong;Hao, Peng;Wang, Zhaoyang;Li, Xuwen;Wang, Zhaoyang

作者机构：

关键词： Animals; Skeleton; Joints; Data models; Predictive models; Feature extraction; Computational modeling; Measurement; Accuracy; Three-dimensional displays; Animal skeletons; keypoint detection; animal pose estimation; feature extraction

期刊名称：IEEE ACCESS （ 影响因子：3.6； 五年影响因子：3.9 ）

ISSN： 2169-3536

年卷期： 2025 年 13 卷

页码：

收录情况： SCI

摘要： Research on two-dimensional keypoint detection within the domain of computer vision has experienced significant advancements. In contrast to the high precision and applicability achieved in human applications, the field of animal keypoint detection remains in its nascent stages of development. To explore the applications and potential of skeleton keypoint detection in areas such as animal pose estimation, behavior recognition, and intelligent breeding. To this end, we focuse on analyzing animal skeletal keypoint detection models based on deep learning technology, examining their keypoint representation forms and multi-object detection strategies. Additionally, it provides a detailed analysis of the method's practical application details and improvements. The paper not only summarizes different model algorithms, datasets, and evaluation metrics related to animal keypoint detection but also integrates various application scenarios, highlighting distinct features under different focal points. Ultimately, this review is expected to broaden the research horizons and methodologies related to animal intelligent behavior recognition, animal welfare studies, and intelligent breeding among scholars.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序