Talking Head Generation
Talking head generation is the task of generating a talking face from a set of images of a person.
( Image credit: Few-Shot Adversarial Learning of Realistic Neural Talking Head Models )
Papers
Showing 11–20 of 119 papers
All datasetsVoxCeleb2 - 1-shot learningVoxCeleb1 - 1-shot learningVoxCeleb1 - 32-shot learningVoxCeleb1 - 8-shot learningVoxCeleb2 - 8-shot learning100 sleep nights of 8 caregiversVoxCeleb2 - 32-shot learning
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Few-shot Adversarial Model | FID | 48.5 | — | Unverified |
| 2 | CainGAN | FID | 35 | — | Unverified |
| 3 | Fast Bi-layer Avatars (medium size) | CSIM | 0.65 | — | Unverified |
| 4 | First Order Motion Model (medium size) | CSIM | 0.64 | — | Unverified |
| 5 | Few-shot Vid-to-vid (medium size) | CSIM | 0.6 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | X2Face | FID | 45.8 | — | Unverified |
| 2 | Few-shot Adversarial Model | FID | 43 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | X2Face | FID | 56.5 | — | Unverified |
| 2 | Few-shot Adversarial Model | FID | 29.5 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | X2Face | FID | 51.5 | — | Unverified |
| 2 | Few-shot Adversarial Model | FID | 38 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Few-shot Adversarial Model | FID | 42.2 | — | Unverified |
| 2 | CainGAN | FID | 24.9 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Ashok | 10% | 12 | — | Unverified |
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | Few-shot Adversarial Model | FID | 30.6 | — | Unverified |