DeepAI AI Chat
Log In Sign Up

Cascade context encoder for improved inpainting

by   Bartosz Zielinski, et al.

In this paper, we analyze if cascade usage of the context encoder with increasing input can improve the results of the inpainting. For this purpose, we train context encoder for 64x64 pixels images in a standard way and use its resized output to fill in the missing input region of the 128x128 context encoder, both in training and evaluation phase. As the result, the inpainting is visibly more plausible. In order to thoroughly verify the results, we introduce normalized squared-distortion, a measure for quantitative inpainting evaluation, and we provide its mathematical explanation. This is the first attempt to formalize the inpainting measure, which is based on the properties of latent feature representation, instead of L2 reconstruction loss.


page 3

page 4


Comparison of CoModGANs, LaMa and GLIDE for Art Inpainting- Completing M.C Escher's Print Gallery

Digital art restoration has benefited from inpainting models to correct ...

Probabilistic Semantic Inpainting with Pixel Constrained CNNs

Semantic inpainting is the task of inferring missing pixels in an image ...

Context Encoders: Feature Learning by Inpainting

We present an unsupervised visual feature learning algorithm driven by c...

Shift-Net: Image Inpainting via Deep Feature Rearrangement

Deep convolutional networks (CNNs) have exhibited their potential in ima...

Structural inpainting

Scene-agnostic visual inpainting remains very challenging despite progre...

Cahn--Hilliard inpainting with the double obstacle potential

The inpainting of damaged images has a wide range of applications and ma...

Controllable Semantic Image Inpainting

We develop a method for user-controllable semantic image inpainting: Giv...