Results for FlashAvatar
Submission info
| Full name | FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding |
|---|---|
| Publication title | FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding |
| Publication authors | Jun Xiang, Xuan Gao, Yudong Guo, Juyong Zhang |
| Publication venue | CVPR 2024 |
| Publication url | https://arxiv.org/abs/2312.02214 |
| Project link | https://github.com/USTC3DV/FlashAvatar-code |
| Hardware | RTX3080 x 1 |
| Code | Code available with license: MIT |
| Use other data | No |
| Last edited | 24 March, 2025 |
Monocular FLAME Avatar results
Unseen expressions, unseen viewpoints
| PSNR | SSIM | LPIPS | JOD | AKD | CSIM | |
|---|---|---|---|---|---|---|
| Left camera | 14.3197 | 0.6820 | 0.5025 | 3.3665 | 22.1832 | 0.1845 |
| Center camera | 18.3755 | 0.7482 | 0.2927 | 4.7646 | 10.7149 | 0.4862 |
| Right camera | 16.2056 | 0.7625 | 0.3614 | 4.3071 | 24.6384 | 0.2403 |
| Total | 16.3002 | 0.7309 | 0.3855 | 4.1461 | 19.1789 | 0.3037 |
Unseen expressions, train viewpoint
| PSNR | SSIM | LPIPS | JOD | AKD | CSIM | |
|---|---|---|---|---|---|---|
| Train viewpoint | 20.6828 | 0.7776 | 0.2158 | 5.4150 | 7.3456 | 0.5997 |
Generated Images
| Subject ID | Left camera | Center camera | Right camera | Train camera | Ground truth |
|---|---|---|---|---|---|
| 393 |
|
|
|
|
|
| 404 |
|
|
|
|
|
| 461 |
|
|
|
|
|
| 477 |
|
|
|
|
|
| 486 |
|
|
|
|
|