Abstract: Leaves of Camphora longepaniculata can be used to extract essential oils, which possess various functions such as antimicrobial, insecticidal, anti-inflammatory, and antioxidant properties. These oils are widely applied in the pharmaceutical, food, and personal care industries. The accurate estimation of leaf fresh mass is a key step for yield prediction and quality control. However, to date, there is no studies on using depth camera to obtain 3D point cloud data for estimating it. In this study, we proposed a method for accurately estimating the leaf fresh mass of C. longepaniculata trees in Yibin City using depth cameras and 3D point cloud technology. By deploying three depth cameras to capture 360-degree 3D point cloud images of individual trees, we extracted key factors, including tree height, canopy width, volume, convex hull surface area, and convex hull volume. These factors were then used to develop a leaf fresh mass estimation model via a multiple linear regression algorithm. To address the issue of limited data, we applied a data augmentation strategy to expand the sample size and enhance the model's robustness. Ten-fold cross-validation results showed that the model achieved an R² of 92.77%, with mean absolute error (MAE) and root mean square error (RMSE) values of 0.160 kg and 0.207 kg, respectively, demonstrating high prediction accuracy. In view of the errors in the model, it suggested that future fusion should adopt multi-source sensor technology and optimize 3D data processing, so as to realize the accurate management and sustainable development of C. longepaniculata industry in the entire lifecycle.