FSMC-Pose: Frequency and Spatial Fusion with Multiscale Self-calibration for Cattle Mounting Pose Estimation

2026-03-17

AI Summary

This work addresses the challenging problem of cow mounting pose estimation in real-world scenarios characterized by complex backgrounds and frequent occlusions. To this end, we propose the FSMC-Pose framework, which comprises a frequency–spatial fused backbone network, CattleMountNet—integrating a Spatial-Frequency Enhancement Block (SFEBlock) and a Receptive Field Aggregation Block (RABlock)—along with a multi-scale spatial-channel self-calibration head (SC²Head). This architecture effectively disentangles targets from cluttered backgrounds, captures multi-scale contextual information, and mitigates structural misalignment caused by occlusions. Evaluated on the newly curated MOUNT-Cattle dataset (conforming to COCO format) and jointly trained with NWAFU-Cattle, our method outperforms strong baselines in accuracy while significantly reducing model parameters and computational cost, enabling real-time inference on commercial GPUs.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

Mounting posture is an important visual indicator of estrus in dairy cattle. However, achieving reliable mounting pose estimation in real-world environments remains challenging due to cluttered backgrounds and frequent inter-animal occlusion. We present FSMC-Pose, a top-down framework that integrates a lightweight frequency-spatial fusion backbone, CattleMountNet, and a multiscale self-calibration head, SC2Head. Specifically, we design two algorithmic components for CattleMountNet: the Spatial Frequency Enhancement Block (SFEBlock) and the Receptive Aggregation Block (RABlock). SFEBlock separates cattle from cluttered backgrounds, while RABlock captures multiscale contextual information. The Spatial-Channel Self-Calibration Head (SC2Head) attends to spatial and channel dependencies and introduces a self-calibration branch to mitigate structural misalignment under inter-animal overlap. We construct a mounting dataset, MOUNT-Cattle, covering 1176 mounting instances, which follows the COCO format and supports drop-in training across pose estimation models. Using a comprehensive dataset that combines MOUNT-Cattle with the public NWAFU-Cattle dataset, FSMC-Pose achieves higher accuracy than strong baselines, with markedly lower computational and parameter costs, while maintaining real-time inference on commodity GPUs. Extensive experiments and qualitative analyses show that FSMC-Pose effectively captures and estimates cattle mounting pose in complex and cluttered environments. Dataset and code are available at https://github.com/elianafang/FSMC-Pose.

Problem

• cattle mounting pose estimation
• cluttered backgrounds
• inter-animal occlusion
• estrus detection
• pose estimation in complex environments

Innovation

• frequency-spatial fusion
• multiscale self-calibration
• cattle mounting pose estimation
• occlusion handling
• lightweight pose network