Non-line-of-sight (NLOS) imaging allows for the visualization of hidden scenes around corners through active sensing.Most previous methods for NLOS reconstruction require dense transients acquired through regular scans over a large relay surface, which limits their applicability in realistic scenarios with irregular relay surfaces.In this paper, we propose an unsupervised learning-based framework for NLOS imaging from irregularly undersampled transients (IUT).Our method learns implicit priors from noisy undersampled transients without requiring paired data, which is difficult and expensive to acquire and align. To overcome the ambiguity of measurement consistency constraint in inferring the latent volume, we design a virtual scanning process that enables the network to learn within both range and null spaces for high-quality reconstruction.We devise a physics-guided SURE-based denoiser to enhance robustness to ubiquitous noise in the by-nature low-photon imaging condition. Extensive experiments on both simulated and real-world data validate the performance and generalization of our method.Compared with the state-of-the-art (SOTA) method, our method achieves higher fidelity, greater robustness, and remarkably faster inference times by orders of magnitude.