融合空间与通道重构卷积和注意力的轻量型动物姿态估计

doi:10.3778/j.issn.1002-8331.2311-0057

摘要/Abstract

摘要： 动物姿态估计在行为生态学、动物健康监测、野生动物保护等领域的重要性不断凸显。然而，目前主流的动物姿态估计算法过于关注准确率，导致网络复杂度和计算成本不断攀升，这使得在移动设备和嵌入式平台上的应用受到了限制。针对该问题，提出融合空间与通道重构卷积和金字塔分割注意力的多尺度动物姿态估计网络SPANet。使用金字塔分割注意力与坐标注意力机制，重新设计了高分辨率网络的瓶颈层EPSAneck，在减轻过度使用大卷积核带来的计算成本的同时，增强了网络对有用特征的提取能力；提出了基于空间和通道重构卷积以及坐标注意力机制的SCCAblock基础模块，在显著减少计算冗余和内存访问的同时，增强了通道与空间之间的信息交互；利用反卷积模块对网络输出的特征融合方式进行重新设计，进一步提升了网络的准确率。实验结果表明，提出的网络模型相较于高分辨率网络在AP10K测试集上的平均精度提升了1.8个百分点，同时浮点运算量降低了48.5%、模型参数量减少了67.0%。在AnimalPose数据集上，浮点运算量降低49.5%，模型参数量降低67.0%。实验数据表明，该网络可在降低模型复杂度的同时实现预测精度的小范围提升。

关键词: 动物姿态估计, 轻量型, 高分辨率, 注意力机制, 空间与通道重构卷积

Abstract: The importance of animal pose estimation in fields such as behavioral ecology, animal health monitoring, and wildlife conservation has been increasingly emphasized. However, current mainstream algorithms for animal pose estimation tend to prioritize accuracy, leading to a continuous increase in network complexity and computational cost, which limits their application on mobile devices and embedded platforms. In response to this issue, this paper proposes a multi-scale animal pose estimation network called SPANet, which combines spatial and channel-reconstructing convolutions with pyramid split attention. Firstly, the bottleneck layer EPSAneck of the high-resolution network is redesigned by incorporating pyramid split attention and coordinate attention mechanisms. This redesign not only reduces the computational cost caused by excessive use of large convolutional kernels but also enhances the ability of network to extract useful features. Secondly, the SCCAblock foundational module is introduced, which is based on spatial and channel-reconstructing convolutions as well as coordinate attention mechanisms. This module significantly reduces computational redundancy and memory access while enhancing information exchange between channels and spatial dimensions. Lastly, the fusion method of network output features is re-designed using deconvolution modules to further improve the accuracy of the network. Experimental results demonstrate that compared to the high-resolution network, the proposed network model achieves an average precision improvement of 1.8 percentage points on the AP10K test set, while reducing the floating-point operations by 48.5% and the number of model parameters by 67.0%. On the AnimalPose dataset, the floating-point operations are reduced by 49.5%, and the number of model parameters is reduced by 67.0%. The experimental data indicate that the proposed network model achieves a small-range improvement in prediction accuracy while reducing the complexity of the model.

Key words: animal pose estimation, lightweight, high-resolution, attention mechanism, spatial and channel reconstruction convolution

宰清鹏, 徐杨. 融合空间与通道重构卷积和注意力的轻量型动物姿态估计[J]. 计算机工程与应用, 2025, 61(6): 282-294.

ZAI Qingpeng, XU Yang. Lightweight Animal Pose Estimation with Integrated Spatial and Channel Reconstructive Convolutions and Attention[J]. Computer Engineering and Applications, 2025, 61(6): 282-294.

参考文献

[1] 吴赛赛, 吴建寨, 程国栋, 等.基于姿态估计的动物行为识别研究进展[J].中国农业大学学报, 2023, 28(6):22-35.
WU S S, WU J Z, CHENG G D, et al. Research progress of animal behavior recognition based on pose estimation[J]. Journal of China Agricultural University, 2023, 28(6):22-35.
[2] 申通, 王硕, 李孟, 等.深度学习在动物行为分析中的应用研究进展[J].计算机科学与探索, 2024, 18(3): 612-626.
SHEN T, WANG S, LI M, et al. Research progress in application of deep learning in animal behavior analysis[J]. Journal of Frontiers of Computer Science and Technology, 2024, 18(3): 612-626.
[3] JIANG L, LEE C, TEOTIA D, et al. Animal pose estimation: a closer look at the state-of-the-art, existing gaps and opportunities[J]. Computer Vision and Image Understanding, 2022, 222:103483.
[4] 张宏鸣, 汪润, 董佩杰, 等.基于DeepSORT算法的肉牛多目标跟踪方法[J].农业机械学报, 2021, 52(4):248-256.
ZHANG H M, WANG R, DONG P J, et al. Beef cattle multi-target tracking based on DeepSORT algorithm[J].Transactions of the Chinese Society for Agricultural Machinery, 2021, 52(4): 248-256.
[5] 徐贵冬, 徐杨, 邓辉, 等.改进高分辨率网络的多目标动物姿态估计研究[J].计算机工程与应用, 2023, 59(22):182-192.
XU G D, XU Y, DENG H, et al. Research on multi-target animal pose estimation based on improved high resolution network[J].Computer Engineering and Applications, 2023, 59(22):182-192.
[6] NEWELL A, YANG K, DENG J. Stacked hourglass networks for human pose estimation[C]//Proceedings of the 14th European Conference on Computer Vision (ECCV 2016), Amsterdam, 2016: 483-499.
[7] CHEN Y, WANG Z, PENG Y, et al. Cascaded pyramid network for multi-person pose estimation[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7103-7112.
[8] SANDLER M, HOWARD A, ZHU M, et al. MobileNetV2: inverted residuals and linear bottlenecks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018:4510-4520.
[9] CAO Z, SIMON T, WEI S E, et al. Realtime multi-person 2D pose estimation using part affinity fields[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017: 7291-7299.
[10] ZHANG Z, TANG J, WU G. Simple and lightweight human pose estimation[J]. arXiv:1911.10346, 2019.
[11] SUN K, XIAO B, LIU D, et al. Deep high-resolution representation learning for human pose estimation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 5693-5703.
[12] CHENG B, XIAO B, WANG J, et al. HigherHRNet: scale-aware representation learning for bottom-up human poseestimation[C]//Proceedings of the IEEE/CVF Conferenceon Computer Vision and Pattern Recognition, 2020: 5386-5395.
[13] YU C, XIAO B, GAO C, et al. Lite-HRNet: a lightweight high-resolution network[C]//Proceedings of the IEEE/CVF Conference on Computervision and Pattern Recognition, 2021: 10440-10450.
[14] ZHANG H, ZU K, LU J, et al. EPSANet: an efficient pyramid squeeze attention block on convolutional neuralnetwork[C]//Proceedings of the Asian Conference on Computer Vision, 2022: 1161-1177.
[15] LI J, WEN Y, HE L. SCConv: spatial and channel reconstruction convolution for feature redundancy[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023: 6153-6162.
[16] TOSHEV A, SZEGEDY C. DeepPose: human pose estimation via deep neural networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2014:1653-1660.
[17] HU J, SHEN L, SUN G. Squeeze-and-excitation networks[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[18] WOO S, PARK J, LEE J Y, et al. CBAM: convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018: 3-19.
[19] WANG Q, WU B, ZHU P, et al. ECA-Net: efficient channel attention for deep convolutional neural networks[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 11534-11542.
[20] LIU H, LIU F, FAN X, et al. Polarized self-attention: towards high-quality pixel-wise regression[J]. arXiv:2107.00782, 2021.
[21] HOU Q, ZHOU D, FENG J. Coordinate attention for efficient mobile network design[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 13713-13722.
[22] LI Y, CHEN Y, DAI X, et al. MicroNet: improving image recognition with extremely low FLOPs[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 468-477.
[23] SINGH P, VERMA V K, RAI P, et al. HetConv: heterogeneous kernel-based convolutions for deep CNNs[C]//Proceedings of the IEEE/CVF Conference on Computer Visionand Pattern Recognition, 2019: 4835-4844.
[24] ZHANG Q, JIANG Z, LU Q, et al. Split to be slim: an overlooked redundancy in vanilla convolution[J]. arXiv:2006. 12085, 2020.
[25] HAN K, WANG Y, TIAN Q, et al. GhostNet: more features from cheap operations[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 1580-1589.
[26] QIU J, CHEN C, LIU S, et al. SlimConv:reducing channelredundancy in convolutional neural networks by features recombining[J].IEEE Transactions on Image Processing, 2021, 30:6434-6445.
[27] CHEN Y, FAN H, XU B, et al. Drop an octave:reducingspatial redundancy in convolutional neural networks with octave convolution[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019: 3435-3444.
[28] IANDOLA F N, HAN S, MOSKEWICZ M W, et al.SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and<0.5 MB model size[J]. arXiv:1602.07360, 2016.
[29] ZHANG X, ZHOU X, LIN M, et al. ShuffleNet: an extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 6848-6856.
[30] HOWARD A G, ZHU M, CHEN B, et al. MobileNets: efficient convolutional neural networks for mobile vision applications[J]. arXiv:1704.04861, 2017.
[31] YU H, XU Y, ZHANG J, et al. AP-10K:a benchmark for animal pose estimation in the wild[J]. arXiv:2108.12617, 2021.
[32] CAO J, TANG H, FANG H S, et al. Cross-domain adaptation for animal pose estimation[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019:9498-9507.
[33] 邓辉, 徐杨.融入注意力和密集连接的轻量型人体姿态估计[J].计算机工程与应用, 2022, 58(16):265-273.
DENG H, XU Y. Lightweight human pose estimation based on attention and dense connection[J].Computer Engineering and Applications, 2022, 58(16):265-273.
[34] XIAO B, WU H, WEI Y. Simple baselines for human pose estimation and tracking[C]//Proceedings of the European Conference on Computer Vision (ECCV), 2018:466-481.
[35] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[36] MA N, ZHANG X, ZHENG H T, et al. ShuffleNetv2: practical guidelines for efficient cnn architecture design[C]//Proceedings of the European Conference on Computer Vision(ECCV), 2018: 116-131.

编辑推荐 0

Metrics

阅读次数

全文

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	0	0	0	29

	来源	本网站

	次数	29
	比例	100%

摘要

最新录用	在线预览	正式出版

0	0	19

	来源	本网站

	次数	19
	比例	100%