Research map / Point clouds / Shape and surface completion
Led3D: A Lightweight and Efficient Deep Approach to Recognizing Low-Quality 3D Faces
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 |
|---|---|---|---|
| Add MSFF to the basic network (ablation) | Led3D · Lock3DFace | rank-one recognition rate (%): 64.88 -> 67.30 | Helped |
| Add SAV to the basic network (ablation) | Led3D · Lock3DFace | rank-one recognition rate (%): 64.88 -> 76.09 | Helped |
| Use both MSFF and SAV (ablation) | Led3D · Lock3DFace | rank-one recognition rate (%): 64.88 -> 81.02 | Helped |
| Augment training data | Led3D · Lock3DFace, average across test subsets | rank-one score: 41.65 -> 54.28 | Helped |
| Concatenate depth and normal features | Led3D · Lock3DFace, total | rank-one recognition rate (%): 81.02 -> 84.22 | Helped |
| Use the proposed network instead of Inception V2 | 3D face recognition · Bosphorus, HL | rank-one recognition rate (%): 78.56 -> 91.27 | Helped |
| Use the proposed network instead of Inception V2 | 3D face recognition · Bosphorus, LL | rank-one recognition rate (%): 77.23 -> 90.70 | Helped |
Closest papers by meaning
- Learning From Millions of 3D Scans for Large-Scale 3D Face Recognition (CVPR 2018 · similarity 0.91)
- Regressing Robust and Discriminative 3D Morphable Models With a Very Deep Neural Network (CVPR 2017 · similarity 0.89)
- FaceScape: A Large-Scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction (CVPR 2020 · similarity 0.87)
- Enhancing Face Recognition With Self-Supervised 3D Reconstruction (CVPR 2022 · similarity 0.87)
- Learning Flexibly Distributional Representation for Low-quality 3D Face Recognition (AAAI 2021 · similarity 0.86)
- Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space (CVPR 2025 · similarity 0.85)
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.