In incremental forming sheet metal is shaped into a desired geometry through a series of deformation steps. This process is advantageous for its flexibility in producing complex parts without requiring dedicated dies. Springback resulting from elastic recovery of the material is the major cause of geometric inaccuracy in the resulting shape from this process. Accurate prediction of springback and controlling deformations are crucial for achieving desired part geometries and dimensional accuracy. This study proposes a Gaussian Process-based approach for springback prediction of the target geometry, utilizing advanced techniques like multitask regression and radial basis function kernels. The training data for the model is generated by finite element simulation of the process. The assessment of the proposed method indicates that this approach enhances the understanding of deformation mechanisms in sheet metal during incremental forming and provides a systematic means to refine trajectories for improved part accuracy and quality.