Disentangling Mean Embeddings for Better Diagnostics of Image Generators
Abstract
The evaluation of image generators remains a challenge due to the limitations of traditional metrics in providing nuanced insights into specific image regions. This is a critical problem as not all regions of an image may be learned with similar ease. In this work, we propose a novel approach to disentangle the cosine similarity of mean embeddings into the product of cosine similarities for individual pixel clusters via central kernel alignment. Consequently, we can quantify the contribution of the cluster-wise performance to the overall image generation performance. We demonstrate how this enhances the explainability and the likelihood of identifying pixel regions of model misbehavior across various real-world use cases.
Cite
Text
Gruber et al. "Disentangling Mean Embeddings for Better Diagnostics of Image Generators." NeurIPS 2024 Workshops: InterpretableAI, 2024.Markdown
[Gruber et al. "Disentangling Mean Embeddings for Better Diagnostics of Image Generators." NeurIPS 2024 Workshops: InterpretableAI, 2024.](https://mlanthology.org/neuripsw/2024/gruber2024neuripsw-disentangling/)BibTeX
@inproceedings{gruber2024neuripsw-disentangling,
title = {{Disentangling Mean Embeddings for Better Diagnostics of Image Generators}},
author = {Gruber, Sebastian Gregor and Ziegler, Pascal Tobias and Buettner, Florian},
booktitle = {NeurIPS 2024 Workshops: InterpretableAI},
year = {2024},
url = {https://mlanthology.org/neuripsw/2024/gruber2024neuripsw-disentangling/}
}