mlsanityData to 7 Oct 2026

Spectrum AUC Difference (SAUCD): Human-aligned 3D Shape Evaluation

arXiv 2403.01619 ↗First posted 3 Mar 2024cs.CV cs.GRcited by 9

Not on the map. Why?

Abstract (arXiv)

Existing 3D mesh shape evaluation metrics mainly focus on the overall shape but are usually less sensitive to local details. This makes them inconsistent with human evaluation, as human perception cares about both overall and detailed shape. In this paper, we propose an analytic metric named Spectrum Area Under the Curve Difference (SAUCD) that demonstrates better consistency with human evaluation. To compare the difference between two shapes, we first transform the 3D mesh to the spectrum domain using the discrete Laplace-Beltrami operator and Fourier transform. Then, we calculate the Area Under the Curve (AUC) difference between the two spectrums, so that each frequency band that captures either the overall or detailed shape is equitably considered. Taking human sensitivity across frequency bands into account, we further extend our metric by learning suitable weights for each frequency band which better aligns with human perception. To measure the performance of SAUCD, we build a 3D mesh evaluation dataset called Shape Grading, along with manual annotations from more than 800 subjects. By measuring the correlation between our metric and human evaluation, we demonstrate that SAUCD is well aligned with human evaluation, and outperforms previous 3D mesh metrics.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 34 of its 83 references that are in the map. First posted on arXiv 3 Mar 2024.

Builds on

34 of 83 references.

References · 34

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2018Pixel2Mesh: Generating 3D Mesh Models from Single RGB Imagescited by 649
  3. 2019Convolutional Mesh Regression for Single-Image Human Shape Reconstructioncited by 270
  4. 2019What Do Single-view 3D Reconstruction Networks Learn?cited by 259
  5. 2020Fast Differentiable Sorting and Rankingcited by 196
  6. 20193D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutionscited by 173
  7. 2020Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes from a Single Imagecited by 170
  8. 2022FaceVerse: a Fine-grained and Detail-controllable 3D Face Morphable Model from a Hybrid Datasetcited by 98
  9. 2021Enriching Local and Global Contexts for Temporal Action Localizationcited by 64
  10. 2020Multimodal Shape Completion via Conditional Generative Adversarial Networkscited by 61
  11. 2020Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localizationcited by 53
  12. 2020DeepHandMesh: A Weakly-supervised Deep Encoder-Decoder Framework for High-fidelity Hand Mesh Modelingcited by 49
22 more
  1. 2020DECOR-GAN: 3D Shape Detailization by Conditional Refinementcited by 45
  2. 2022Learning to Generate 3D Shapes from a Single Examplecited by 31
  3. 2022Min-Max Similarity: A Contrastive Semi-Supervised Deep Learning Network for Surgical Tools Segmentationcited by 28
  4. 2021PC-HMR: Pose Calibration for 3D Human Mesh Recovery from 2D Images/Videoscited by 25
  5. 2020Learning Delaunay Surface Elements for Mesh Reconstructioncited by 23
  6. 2023Towards Generic Image Manipulation Detection with Weakly-Supervised Self-Consistency Learningcited by 23
  7. 2022Point Scene Understanding via Disentangled Instance Mesh Reconstructioncited by 20
  8. 2023Cycle-guided Denoising Diffusion Probability Model for 3D Cross-modality MRI Synthesiscited by 18
  9. 2023High Fidelity 3D Hand Shape Reconstruction via Scalable Graph Frequency Decompositioncited by 18
  10. 2023Language-guided Human Motion Synthesis with Atomic Actionscited by 17
  11. 2021Deep Hybrid Self-Prior for Full 3D Mesh Generationcited by 13
  12. 2018Learning to Generate and Reconstruct 3D Meshes with only 2D Supervisioncited by 12
  13. 2021Task-Oriented Low-Dose CT Image Denoisingcited by 11
  14. 2021Fostering Generalization in Single-view 3D Reconstruction by Learning a Hierarchy of Local and Global Shape Priorscited by 9
  15. 2022Self-supervised Human Mesh Recovery with Cross-Representation Alignmentcited by 9
  16. 2022Progressive Multi-view Human Mesh Recovery with Self-Supervisioncited by 9
  17. 2020MeshMVS: Multi-View Stereo Guided Mesh Reconstructioncited by 8
  18. 2022Reducing Positional Variance in Cross-sectional Abdominal CT Slices with Deep Conditional Generative Modelscited by 7
  19. 2023SHISRCNet: Super-resolution And Classification Network For Low-resolution Breast Cancer Histopathology Imagecited by 7
  20. 2023Learning Continuous Mesh Representation with Spherical Implicit Surfacecited by 3
  21. 2022CD2: Fine-grained 3D Mesh Reconstruction With Twice Chamfer Distancecited by 1
  22. 2023Deep conditional generative models for longitudinal single-slice abdominal computed tomography harmonizationcited by 1

Built on by

Cited by year

Citing papers per year: 2024: 6; 2025: 3.

9 papers.

  1. 2024MH-pFLID: Model Heterogeneous personalized Federated Learning via Injection and Distillation for Medical Data Analysiscited by 14
  2. 2024IDOL: Unified Dual-Modal Latent Diffusion for Human-Centric Joint Video-Depth Generationcited by 11
  3. 2024AnatoMask: Enhancing Medical Image Segmentation with Reconstruction-guided Self-maskingcited by 10
  4. 2025dFLMoE: Decentralized Federated Learning via Mixture of Experts for Medical Data Analysiscited by 8
  5. 2024Disease-informed Adaptation of Vision-Language Modelscited by 7
  6. 2024FSC: Few-point Shape Completioncited by 6
  7. 2024Divide and Fuse: Body Part Mesh Recovery from Partially Visible Human Imagescited by 1
  8. 2025SRAM: Shape-Realism Alignment Metric for No Reference 3D Shape Evaluationcited by 0
  9. 2025Textured Geometry Evaluation: Perceptual 3D Textured Shape Metric via 3D Latent-Geometry Networkcited by 0