Abstract | ||
---|---|---|
Our work focuses on addressing sample deficiency from low-density regions of data manifold in common image datasets. We leverage diffusion process based generative models to synthesize novel images from low-density regions. We observe that uniform sampling from diffusion models predominantly samples from high-density regions of the data manifold. Therefore, we modify the sampling process to guide it towards low-density regions while simulta-neously maintaining the fidelity of synthetic data. We rigorously demonstrate that our process successfully generates novel high fidelity samples from low-density regions. We further examine generated samples and show that the model does not memorize low-density data and indeed learns to generate novel samples from low-density regions. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/CVPR52688.2022.01120 | IEEE Conference on Computer Vision and Pattern Recognition |
Keywords | DocType | Volume |
Image and video synthesis and generation, Machine learning, Representation learning | Conference | 2022 |
Issue | Citations | PageRank |
1 | 0 | 0.34 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Vikash Sehwag | 1 | 17 | 5.32 |
Caner Hazirbas | 2 | 0 | 0.34 |
Albert Gordo | 3 | 0 | 1.69 |
Firat Ozgenel | 4 | 0 | 0.34 |
Cristian Canton Ferrer | 5 | 0 | 0.34 |