This article reviews computer vision-based techniques for non-contact measurement of livestock body dimensions and weight, highlighting their advantages over traditional manual methods. It discusses the integration of next-generation AI, visual processing, intelligent sensing, multimodal fusion, and robotic technology in livestock farming. The review focuses on three main areas: 3D reconstruction of livestock, body dimension acquisition using computer vision, and weight estimation via computer vision. The article analyzes various methods for 3D reconstruction, including those based on RGB images, laser scanning, and 3D cameras, and compares their effectiveness. It also explores body size calculation methods, comparing RGB image-based and 3D point cloud-based approaches. Weight estimation methods, including linear regression and neural networks, are discussed, along with the challenges and future trends in non-contact livestock phenotypic data acquisition. The review emphasizes the importance of non-contact methods in precision livestock farming (PLF) for improving data collection efficiency, accuracy, and animal welfare. Key challenges include the lack of accurate 3D reconstruction models, limited high-quality livestock point cloud datasets, inefficient data acquisition methods, and high cost-effectiveness. Future research directions include enhancing neural network model accuracy, standardizing data, and improving data collection efficiency. The article concludes that non-contact computer vision-based techniques are becoming the primary direction for acquiring livestock phenotypic data, with the potential for significant advancements in precision livestock farming.This article reviews computer vision-based techniques for non-contact measurement of livestock body dimensions and weight, highlighting their advantages over traditional manual methods. It discusses the integration of next-generation AI, visual processing, intelligent sensing, multimodal fusion, and robotic technology in livestock farming. The review focuses on three main areas: 3D reconstruction of livestock, body dimension acquisition using computer vision, and weight estimation via computer vision. The article analyzes various methods for 3D reconstruction, including those based on RGB images, laser scanning, and 3D cameras, and compares their effectiveness. It also explores body size calculation methods, comparing RGB image-based and 3D point cloud-based approaches. Weight estimation methods, including linear regression and neural networks, are discussed, along with the challenges and future trends in non-contact livestock phenotypic data acquisition. The review emphasizes the importance of non-contact methods in precision livestock farming (PLF) for improving data collection efficiency, accuracy, and animal welfare. Key challenges include the lack of accurate 3D reconstruction models, limited high-quality livestock point cloud datasets, inefficient data acquisition methods, and high cost-effectiveness. Future research directions include enhancing neural network model accuracy, standardizing data, and improving data collection efficiency. The article concludes that non-contact computer vision-based techniques are becoming the primary direction for acquiring livestock phenotypic data, with the potential for significant advancements in precision livestock farming.