# Gram-MMD: A Texture-Aware Metric for Image Realism Assessment

arXiv: https://arxiv.org/abs/2604.03064 · HTML: https://mlsanity.com/paper/2604.03064/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 29 references that are in the map.
- First posted on arXiv 3 Apr 2026.

## Builds on (most cited first, 18 of 18)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Visualizing and Understanding Convolutional Networks](https://mlsanity.com/paper/1311.2901/) (2013)
- [Demystifying MMD GANs](https://mlsanity.com/paper/1801.01401/) (2018)
- [Do Vision Transformers See Like Convolutional Neural Networks?](https://mlsanity.com/paper/2108.08810/) (2021)
- [A Neural Algorithm of Artistic Style](https://mlsanity.com/paper/1508.06576/) (2015)
- [Virtual Worlds as Proxy for Multi-Object Tracking Analysis](https://mlsanity.com/paper/1605.06457/) (2016)
- [Rethinking FID: Towards a Better Evaluation Metric for Image Generation](https://mlsanity.com/paper/2401.09603/) (2023)
- [Deep Compression Autoencoder for Efficient High-Resolution Diffusion Models](https://mlsanity.com/paper/2410.10733/) (2024)
- [On Riemannian Optimization over Positive Definite Matrices with the Bures-Wasserstein Geometry](https://mlsanity.com/paper/2106.00286/) (2021)
- [Kernel Methods on the Riemannian Manifold of Symmetric Positive Definite Matrices](https://mlsanity.com/paper/1412.4172/) (2014)
- [How good are humans at detecting AI-generated images? Learnings from an experiment](https://mlsanity.com/paper/2507.18640/) (2025)

## Built on by (most cited first, 1 of 1)

- [TRIG: Toward Measuring Texture Realism in Image Generation](https://mlsanity.com/paper/2610.15164/) (2026)

## Abstract (arXiv)

Evaluating the realism of generated images remains a fundamental challenge in generative modeling. Existing distributional metrics such as the Frechet Inception Distance (FID) and CLIP-MMD (CMMD) compare feature distributions at a semantic level but may overlook fine-grained textural information that can be relevant for distinguishing real from generated images. We introduce Gram-MMD (GMMD), a realism metric that leverages Gram matrices computed from intermediate activations of pretrained backbone networks to capture correlations between feature maps. By extracting the upper-triangular part of these symmetric Gram matrices and measuring the Maximum Mean Discrepancy (MMD) between an anchor distribution of real images and an evaluation distribution, GMMD produces a representation that encodes textural and structural characteristics at a finer granularity than global embeddings. To select the hyperparameters of the metric, we employ a meta-metric protocol based on controlled degradations applied to MS-COCO images, measuring monotonicity via Spearman's rank correlation and Kendall's tau. We conduct experiments on both the KADID-10k database and the RAISE realness assessment dataset using various backbone architectures, including DINOv2, DC-AE, Stable Diffusion's VAE encoder, VGG19, and the AlexNet backbone from LPIPS, among others. We also demonstrate on a cross-domain driving scenario (KITTI / Virtual KITTI / Stanford Cars) that CMMD can incorrectly rank real images as less realistic than synthetic ones due to its semantic bias, while GMMD preserves the correct ordering. Our results suggest that GMMD captures complementary information to existing semantic-level metrics.