REALY (Region-aware benchmark based on the LYHM)
Introduced by Chai et al. in REALY: Rethinking the Evaluation of 3D Face Reconstruction
The REALY benchmark aims to introduce a region-aware evaluation pipeline to measure the fine-grained normalized mean square error (NMSE) of 3D face reconstruction methods from under-controlled image sets.
Given the reconstructed mesh from the 2D image in REALY by a specific method, the REALY benchmark calculates the similarity of ground-truth scans on four regions (nose, mouth, forehead, cheek) with the predicted mesh.
Benchmarks
| Trend | Task | Dataset Variant | Best Model | Paper | Code |
|---|---|---|---|---|---|
|
REALY
|
HiFace-f
|
|||
|
REALY (side-view)
|
HiFace-f
|
Papers
| Paper | Code | Results | Date | Stars |
|---|
