Research map / 3D vision / Face and shape
Adaptive 3D Face Reconstruction From Unconstrained Photo Collections
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · Synthetic neutral frontal | normalized vertex error: 6.13% -> 3.22% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · Synthetic neutral faces, random yaw | normalized vertex error: 7.48% -> 3.82% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · Synthetic frontal faces, random expression | normalized vertex error: 6.59% -> 4.40% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · PC1 | normalized vertex error: 8.31% -> 5.10% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · PC2, 40 photos | normalized vertex error: 4.13% -> 3.18% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · PC2, 20 photos | normalized vertex error: 4.73% -> 3.46% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · PC2, 10 photos | normalized vertex error: 5.40% -> 4.03% | Helped |
| Use the adaptive personalized-template reconstruction instead of [28] | 3D face reconstruction · PC2, 5 photos | normalized vertex error: 8.77% -> 4.07% | Helped |
| Increase the photo collection from 1 to 40 images (ablation) | Adaptive reconstruction · PC2 | normalized vertex error: 4.19% -> 3.18% | Helped |
| Remove coarse-to-fine reconstruction and start at fine resolution (ablation) | Adaptive reconstruction · George Clooney collection | processing time: 3.5 minutes -> 7 minutes | Hurt |
| Remove coarse-to-fine reconstruction and start at fine resolution (ablation) | Adaptive reconstruction · Small photo collections | reconstruction quality | Hurt |
| Add camera-facing dependability weights to image contributions (ablation) | Photometric normal and albedo estimation · Face reconstruction | albedo estimation quality | Helped |
Closest papers by meaning
- Sparse Photometric 3D Face Reconstruction Guided by Morphable Models (CVPR 2018 · similarity 0.91)
- Learning Complete 3D Morphable Face Models From Images and Videos (CVPR 2021 · similarity 0.90)
- A Morphable Face Albedo Model (CVPR 2020 · similarity 0.88)
- Lightweight Photometric Stereo for Facial Details Recovery (CVPR 2020 · similarity 0.88)
- Pixel3DMM: Versatile Screen-Space Priors for Single-Image 3D Face Reconstruction (ICLR 2026 · similarity 0.88)
- PSHuman: Photorealistic Single-image 3D Human Reconstruction using Cross-Scale Multiview Diffusion and Explicit Remeshing (CVPR 2025 · similarity 0.86)
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.