LPIPS metric. pip install lpips

A metric for Perceptual Image-Error Assessment through Pairwise Preference (PieAPP at CVPR 2018).

A simple and useful implementation of LPIPS.

Comparing different similarity functions for reconstruction of image on CycleGAN. (https://tandon-a.github.io/CycleGAN_ssim/) Training cycleGAN with different loss functions to improve visual quality of produced images

[ECCV 2022] We investigated a broad range of neural network elements and developed a robust perceptual similarity metric. Our shift-tolerant perceptual similarity metric (ST-LPIPS) is consistent with human perception and is less susceptible to imperceptible misalignments between two images than existing metrics.

Experiments with perceptual loss and autoencoders.

A no-reference version of HDR-VDP using deep-learning

LPIPS metric on PaddlePaddle. pip install paddle-lpips

Benchmarking library for image manipulation detection.

A Study of Deep Perceptual Metrics for Image quality Assessment

[TMLR 2023] as a featured article (spotlight 🌟 or top 0.01% of the accepted papers). In this study, we systematically examine the robustness of both traditional and learned perceptual similarity metrics to imperceptible adversarial perturbations.

Official implementation of the paper: Learning a distance measure from the information-estimation geometry of data

Finetuning and clustering library for image perceptual similarity models.

Android librarry (kotlin) : Image (JPEG, BMP) comparison (perceptual hash algorithm)

Clustering slices within NIfTI volume based on Perceptual Similarity or SSIM.

An evaluation of popular optimisation-based and encoder-based GAN inversion (image to latent code) approaches. The project utilises the StyleGAN2-ada architecture and for ease of reproduction has been performed on the publically available FFHQ dataset with pre-trained weights.