Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2411.12002 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912124030156800 |
|---|---|
| author | Zeng, Libing Kalantari, Nima Khademi |
| author_facet | Zeng, Libing Kalantari, Nima Khademi |
| contents | With the advances in generative adversarial networks (GANs) and neural rendering, 3D relightable face generation has received significant attention. Among the existing methods, a particularly successful technique uses an implicit lighting representation and generates relit images through the product of synthesized albedo and light-dependent shading images. While this approach produces high-quality results with intricate shading details, it often has difficulty producing relit images with consistent skin tones, particularly when the lighting condition is extracted from images of individuals with dark skin. Additionally, this technique is biased towards producing albedo images with lighter skin tones. Our main observation is that this problem is rooted in the biased spherical harmonics (SH) coefficients, used during training. Following this observation, we conduct an analysis and demonstrate that the bias appears not only in band 0 (DC term), but also in the other bands of the estimated SH coefficients. We then propose a simple, but effective, strategy to mitigate the problem. Specifically, we normalize the SH coefficients by their DC term to eliminate the inherent magnitude bias, while statistically align the coefficients in the other bands to alleviate the directional bias. We also propose a scaling strategy to match the distribution of illumination magnitude in the generated images with the training data. Through extensive experiments, we demonstrate the effectiveness of our solution in increasing the skin tone consistency and mitigating bias. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_12002 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Analyzing and Improving the Skin Tone Consistency and Bias in Implicit 3D Relightable Face Generators Zeng, Libing Kalantari, Nima Khademi Computer Vision and Pattern Recognition With the advances in generative adversarial networks (GANs) and neural rendering, 3D relightable face generation has received significant attention. Among the existing methods, a particularly successful technique uses an implicit lighting representation and generates relit images through the product of synthesized albedo and light-dependent shading images. While this approach produces high-quality results with intricate shading details, it often has difficulty producing relit images with consistent skin tones, particularly when the lighting condition is extracted from images of individuals with dark skin. Additionally, this technique is biased towards producing albedo images with lighter skin tones. Our main observation is that this problem is rooted in the biased spherical harmonics (SH) coefficients, used during training. Following this observation, we conduct an analysis and demonstrate that the bias appears not only in band 0 (DC term), but also in the other bands of the estimated SH coefficients. We then propose a simple, but effective, strategy to mitigate the problem. Specifically, we normalize the SH coefficients by their DC term to eliminate the inherent magnitude bias, while statistically align the coefficients in the other bands to alleviate the directional bias. We also propose a scaling strategy to match the distribution of illumination magnitude in the generated images with the training data. Through extensive experiments, we demonstrate the effectiveness of our solution in increasing the skin tone consistency and mitigating bias. |
| title | Analyzing and Improving the Skin Tone Consistency and Bias in Implicit 3D Relightable Face Generators |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2411.12002 |