Rethinking cluster-conditioned diffusion models for label-free image synthesis
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909395292520448 |
|---|---|
| author | Adaloglou, Nikolas Kaiser, Tim Michels, Felix Kollmann, Markus |
| author_facet | Adaloglou, Nikolas Kaiser, Tim Michels, Felix Kollmann, Markus |
| contents | Diffusion-based image generation models can enhance image quality when conditioned on ground truth labels. Here, we conduct a comprehensive experimental study on image-level conditioning for diffusion models using cluster assignments. We investigate how individual clustering determinants, such as the number of clusters and the clustering method, impact image synthesis across three different datasets. Given the optimal number of clusters with respect to image synthesis, we show that cluster-conditioning can achieve state-of-the-art performance, with an FID of 1.67 for CIFAR10 and 2.17 for CIFAR100, along with a strong increase in training sample efficiency. We further propose a novel empirical method to estimate an upper bound for the optimal number of clusters. Unlike existing approaches, we find no significant association between clustering performance and the corresponding cluster-conditional FID scores. The code is available at https://github.com/HHU-MMBS/cedm-official-wavc2025. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_00570 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Rethinking cluster-conditioned diffusion models for label-free image synthesis Adaloglou, Nikolas Kaiser, Tim Michels, Felix Kollmann, Markus Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Diffusion-based image generation models can enhance image quality when conditioned on ground truth labels. Here, we conduct a comprehensive experimental study on image-level conditioning for diffusion models using cluster assignments. We investigate how individual clustering determinants, such as the number of clusters and the clustering method, impact image synthesis across three different datasets. Given the optimal number of clusters with respect to image synthesis, we show that cluster-conditioning can achieve state-of-the-art performance, with an FID of 1.67 for CIFAR10 and 2.17 for CIFAR100, along with a strong increase in training sample efficiency. We further propose a novel empirical method to estimate an upper bound for the optimal number of clusters. Unlike existing approaches, we find no significant association between clustering performance and the corresponding cluster-conditional FID scores. The code is available at https://github.com/HHU-MMBS/cedm-official-wavc2025. |
| title | Rethinking cluster-conditioned diffusion models for label-free image synthesis |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2403.00570 |