Perceptual Autoencoder for Compressive Sensing Image Reconstruction

Ivan Ralašić, Damir Seršić, Siniša Šegvić
2020 Informatica  
This paper presents a non-iterative deep learning approach to compressive sensing (CS) image reconstruction using a convolutional autoencoder and a residual learning network. An efficient measurement design is proposed in order to enable training of the compressive sensing models on normalized and mean-centred measurements, along with a practical network initialization method based on principal component analysis (PCA). Finally, perceptual residual learning is proposed in order to obtain
more » ... er to obtain semantically informative image reconstructions along with high pixel-wise reconstruction accuracy at low measurement rates.
doi:10.15388/20-infor421 fatcat:7orhgmuvcjcthonvzpiucyctcy