Compressive Sensing for Vision - Sparse Representations and Compressive Sensing for Imaging and Vision - page 59

Fig. 4.11 Reconstructed surface when the gradient field is corrupted by both outliers (at 3%
locations) and noise (Gaussian with 7%). (a) Original surface. (b) Image plot of the surface

shown in (a). (c) Surface normal needle plot of the original surface. (d) Surface normal needle plot
corresponding to the noisy gradients. (e) Reconstruction using Poisson solver. (f) Reconstruction

σ =