A Discriminative Distance Metric Learning with Adaptive Label Consistency for Semi-supervised Scene Parsing from Point Clouds

To parse large-scale urban scenes using the supervised methods, a large amount of training data that can account for the vast visual and structural variance of urban environment is necessary. Unfortunately, such training data is mostly obtained by tedious and time-consuming manual work. To overcome the drawback, we propose a semi-supervised learning framework that combines the distance metric learning, co-graph and label constraints into an objective function for point cloud parsing. Mathematically, the distance metric learning is presented to learn a novel distance criterion that can effectively recognize points of different classes. The graph regularization is then employed to characterize the intrinsic geometry structure of the data manifold and explore relationships among points. The label consistency regularization is introduced to ensure the category consistency of the clustered points and single point. To classify the out-of-sample data, the framework successfully transforms the semi-supervised classification results into the linear classifier by adopting a linear regression. An iterative algorithm is utilized to efficiently and effectively optimize the objective function with characteristics of multiple variables and highly nonlinear. The point clouds of four urban scenes are used to validate our method. Experimental results show that our method achieves superior performance about 2 13% higher classification accuracy on the unlabeled data and about 12 25% higher classification accuracy on the test data compared to the state-of-the-art semi-supervised point cloud classification algorithms.