Semantically Consistent Image Completion with Fine-grained Details

11/26/2017
by   Pengpeng Liu, et al.
0

Image completion has achieved significant progress due to advances in generative adversarial networks (GANs). Albeit natural-looking, the synthesized contents still lack details, especially for scenes with complex structures or images with large holes. This is because there exists a gap between low-level reconstruction loss and high-level adversarial loss. To address this issue, we introduce a perceptual network to provide mid-level guidance, which measures the semantical similarity between the synthesized and original contents in a similarity-enhanced space. We conduct a detailed analysis on the effects of different losses and different levels of perceptual features in image completion, showing that there exist complementarity between adversarial training and perceptual features. By combining them together, our model can achieve nearly seamless fusion results in an end-to-end manner. Moreover, we design an effective lightweight generator architecture, which can achieve effective image inpainting with far less parameters. Evaluated on CelebA Face and Paris StreetView dataset, our proposed method significantly outperforms existing methods.

READ FULL TEXT

page 2

page 4

page 5

page 6

page 7

page 8

page 10

page 11

research
12/21/2017

Context-Aware Semantic Inpainting

Recently image inpainting has witnessed rapid progress due to generative...
research
06/14/2021

Context-Aware Image Inpainting with Learned Semantic Priors

Recent advances in image inpainting have shown impressive results for ge...
research
08/06/2018

X-GANs: Image Reconstruction Made Easy for Extreme Cases

Image reconstruction including image restoration and denoising is a chal...
research
04/20/2019

Everyone is a Cartoonist: Selfie Cartoonization with Attentive Adversarial Networks

Selfie and cartoon are two popular artistic forms that are widely presen...
research
02/13/2020

PCSGAN: Perceptual Cyclic-Synthesized Generative Adversarial Networks for Thermal and NIR to Visible Image Transformation

In many real world scenarios, it is difficult to capture the images in t...
research
05/06/2019

The Missing Data Encoder: Cross-Channel Image Completion with Hide-And-Seek Adversarial Network

Image completion is the problem of generating whole images from fragment...
research
08/21/2020

Toward Quantifying Ambiguities in Artistic Images

It has long been hypothesized that perceptual ambiguities play an import...

Please sign up or login with your details

Forgot password? Click here to reset