Regularizing Vector Embedding in Bottom-Up Human Pose Estimation

CASIA OpenIR > 紫东太初大模型研究中心

	Regularizing Vector Embedding in Bottom-Up Human Pose Estimation
	Wang Haixin1,2 ; Zhou Lu2 ; Chen Yingying2 ; Tang Ming1,2 ; Wang Jinqiao1,2,3,4
	2022-10
会议名称	European Conference on Computer Vision
会议日期	October 23-27, 2022
会议地点	Tel Aviv, Israel
摘要	The embedding-based method such as Associative Embed ding is popular in bottom-up human pose estimation. Methods under this framework group candidate keypoints according to the predicted identity embeddings. However, the identity embeddings of different in stances are likely to be linearly inseparable in some complex scenes, such as crowded scene or when the number of instances in the image is large. To reduce the impact of this phenomenon on keypoint grouping, we try to learn a sparse multidimensional embedding for each keypoint. We observe that the different dimensions of embeddings are highly linearly correlated. To address this issue, we impose an additional constraint on the embeddings during training phase. Based on the fact that the scales of instances usually have significant variations, we uilize the scales of in stances to regularize the embeddings, which effectively reduces the linear correlation of embeddings and makes embeddings being sparse. We eval uate our model on CrowdPose Test and COCO Test-dev. Compared to vanilla Associative Embedding, our method has an impressive superiority in keypoint grouping, especially in crowded scenes with a large number of instances. Furthermore, our method achieves state-of-the-art results on CrowdPose Test (74.5 AP) and COCO Test-dev (72.8 AP), outper forming other bottom-up methods. Our code and pretrained models are available at https://github.com/CR320/CoupledEmbedding.
收录类别	EI
七大方向——子方向分类	图像视频处理与分析
国重实验室规划方向分类	视觉信息处理
是否有论文关联数据集需要存交	否
文献类型	会议论文
条目标识符	http://ir.ia.ac.cn/handle/173211/57168
专题	紫东太初大模型研究中心
作者单位	1.School of Artificial Intelligence, University of Chinese Academy of Sciences 2.National Laboratory of Pattern Recognition, Institute of Automation, Chinese Academy of Sciences 3.Peng Cheng Laboratory 4.ObjectEye Inc
第一作者单位	模式识别国家重点实验室
推荐引用方式 GB/T 7714	Wang Haixin,Zhou Lu,Chen Yingying,et al. Regularizing Vector Embedding in Bottom-Up Human Pose Estimation[C],2022.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
Regularizing Vector （1793KB）	会议论文		开放获取	CC BY-NC-SA	浏览下载