Research map / Point clouds / Shape and surface completion
3D Face Modeling From Diverse Raw Scan Data
Feng Liu, Luan Tran, Xiaoming Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add surface normal loss to vertex loss (ablation) | Proposed · Ablation (qualitative) | surface smoothness and local detail | Improved |
| add edge length term to the loss (ablation) | Proposed · Ablation (qualitative) | local triangle topology | Improved |
| train on BU3DFE data (in) instead of outside data (out) | Proposed · BU3DFE (avg over regions) | semantic landmark error (mm): 6.36 -> 5.14 | Improved |
| Proposed (in) vs best baseline GPMMs | Proposed (in) · BU3DFE (avg over regions) | semantic landmark error (mm): 6.52 -> 5.14 | Improved |
| Proposed vs best baseline on mouth region | Proposed (in) · BU3DFE, Mouth | semantic landmark error (mm): 3.66 -> 4.17 | Hurt |
| Proposed vs Fan et al. on FRGC | Proposed · FRGC v2.0 | semantic landmark error (mm): 2.73 -> 2.42 | Improved |
| Proposed vs nonlinear 3DMM at latent dim 40 | Proposed · BFM test scans | per-vertex error (mm): 1.440 -> 1.258 | Improved |
| Proposed vs nonlinear 3DMM at latent dim 160 | Proposed · BFM test scans | per-vertex error (mm): 1.019 -> 0.946 | Improved |
| Proposed vs 3DDFA expression model | Proposed · BFM expression | per-vertex error (mm): 2.609 -> 1.424 | Improved |
| Proposed vs FLAME on COMA | Proposed · COMA | per-vertex error (mm): 1.615 -> 1.474 | Improved |
| Proposed vs MeshAE on COMA | Proposed · COMA | per-vertex error (mm): 1.160 -> 1.474 | Hurt |
Closest papers by meaning
- Generating 3D Faces using Convolutional Mesh Autoencoders (ECCV 2018 · similarity 0.88)
- Learning Dense Facial Correspondences in Unconstrained Images (ICCV 2017 · similarity 0.88)
- A Sparse and Locally Coherent Morphable Face Model for Dense Semantic Correspondence Across Heterogeneous 3D Faces (TPAMI 2021 · similarity 0.88)
- Learning Complete 3D Morphable Face Models From Images and Videos (CVPR 2021 · similarity 0.88)
- Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation (IJCV 2021 · similarity 0.88)
- Dense 3D Face Correspondence (TPAMI 2017 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
