Abstract:
Deep neural networks are now used for image generation and other tasks. GANs can create realistic images for image compression. Like standard compression, generative image perceptual quality should be automatically assessed to monitor and control the encode process.
Existing image quality algorithms fail on GAN-generated content, especially at high compressions and textured regions. We propose a naturalness-based generative image quality predictor.
Our new GAN picture quality predictor uses a multi-stage parallel boosting system based on structural similarity and statistical similarity.
We created a subjective GAN image quality database with distorted GAN images and human ratings to test models. Our GAN IQA model outperforms traditional and generative image quality datasets in our experiments.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
