Supplementary material for " Generalised Implicit Neural Representations "

One interesting application of INRs is to train them using the derivatives of the target signal as supervision. This idea, which was introduced by Sitzmann et al. We report in Figure 1 the original texture, its Laplacian, and the reconstructed signal, as predicted by the INR. In the second experiment of Section 4.1, we train the models using the default setting.