mlsanityData to 7 Oct 2026

NeRD: Neural 3D Reflection Symmetry Detector

arXiv 2105.03211 ↗First posted 19 Apr 2021cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

Recent advances have shown that symmetry, a structural prior that most objects exhibit, can support a variety of single-view 3D understanding tasks. However, detecting 3D symmetry from an image remains a challenging task. Previous works either assume that the symmetry is given or detect the symmetry with a heuristic-based method. In this paper, we present NeRD, a Neural 3D Reflection Symmetry Detector, which combines the strength of learning-based recognition and geometry-based reconstruction to accurately recover the normal direction of objects' mirror planes. Specifically, we first enumerate the symmetry planes with a coarse-to-fine strategy and then find the best ones by building 3D cost volumes to examine the intra-image pixel correspondence from the symmetry. Our experiments show that the symmetry planes detected with our method are significantly more accurate than the planes from direct CNN regression on both synthetic and real-world datasets. We also demonstrate that the detected symmetry can be used to improve the performance of downstream tasks such as pose estimation and depth map regression. The code of this paper has been made public at https://github.com/zhou13/nerd.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 30 of its 48 references that are in the map. First posted on arXiv 19 Apr 2021.

Builds on

30 of 48 references.

References · 30

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  4. 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,186
  5. 2016Stacked Hourglass Networks for Human Pose Estimationcited by 913
  6. 2018Deep Ordinal Regression Network for Monocular Depth Estimationcited by 721
  7. 2018Pixel2Mesh: Generating 3D Mesh Models from Single RGB Imagescited by 649
  8. 2017Frustum PointNets for 3D Object Detection from RGB-D Datacited by 640
  9. 20163D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstructioncited by 587
  10. 2019Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimationcited by 490
  11. 2018Pyramid Stereo Matching Networkcited by 486
  12. 2019Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoningcited by 420
18 more
  1. 2017End-to-End Learning of Geometry and Context for Deep Stereo Regressioncited by 359
  2. 2018GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Posecited by 341
  3. 2019DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstructioncited by 332
  4. 20163D Bounding Box Estimation Using Deep Learning and Geometrycited by 277
  5. 2019What Do Single-view 3D Reconstruction Networks Learn?cited by 259
  6. 2017BB8: A Scalable, Accurate, Robust to Partial Occlusion Method for Predicting the 3D Poses of Challenging Objects without Using Depthcited by 225
  7. 2018Pix3D: Dataset and Methods for Single-Image 3D Shape Modelingcited by 218
  8. 2017Real-Time Seamless Single Shot 6D Object Pose Predictioncited by 210
  9. 2017Learning a Multi-View Stereo Machinecited by 194
  10. 2019Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wildcited by 180
  11. 2018High Quality Monocular Depth Estimation via Transfer Learningcited by 133
  12. 2015Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Imagescited by 111
  13. 2018Unsupervised Learning of Shape and Pose with Differentiable Point Cloudscited by 89
  14. 2020Portrait Shadow Manipulationcited by 42
  15. 2019NeurVPS: Neural Vanishing Point Scanning via Conic Convolutioncited by 24
  16. 2020Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetrycited by 24
  17. 2017Beyond Planar Symmetry: Modeling human perception of reflection and rotation symmetries in the wildcited by 17
  18. 2019Front2Back: Single View 3D Shape Reconstruction via Front to Back Predictioncited by 17

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 4; 2022: 5; 2023: 0; 2024: 3; 2025: 2; 2026: 4.

18 papers.

Sort:

  1. 2025SymmCompletion: High-Fidelity and High-Consistency Point Cloud Completion with Symmetry Guidancecited by 15
  2. 2024Robust Symmetry Detection via Riemannian Langevin Dynamicscited by 14
  3. 2022SymmNeRF: Learning to Explore Symmetry Prior for Single-View View Synthesiscited by 13
  4. 2022SNeS: Learning Probably Symmetric Neural Surfaces from Incomplete Datacited by 9
  5. 2024Symmetry Strikes Back: From Single-Image Symmetry Detection to 3D Generationcited by 7
  6. 2021NeRD++: Improved 3D-mirror symmetry learning from a single imagecited by 4
  7. 2022Learning Proximal Operators to Discover Multiple Optimacited by 4
  8. 2021Black-Box Test-Time Shape REFINEment for Single View 3D Reconstructioncited by 3
  9. 2021Robust Extrinsic Symmetry Estimation in 3D Point Cloudscited by 2
  10. 2022SPA-VAE: Similar-Parts-Assignment for Unsupervised 3D Point Cloud Generationcited by 2
  11. 2021Symmetry Perception by Deep Networks: Inadequacy of Feed-Forward Architectures and Improvements with Recurrent Connectionscited by 1
  12. 2026Quartet of Diffusions: Structure-Aware Point Cloud Generation through Part and Symmetry Guidancecited by 1
6 more
  1. 2022Inferring Implicit 3D Representations from Human Figures on Pictorial Mapscited by 0
  2. 2024Automatic Skull Reconstruction by Deep Learnable Symmetry Enforcementcited by 0
  3. 2025Symmetry Matters: Auditing and Symmetrizing 3D Generative Modelscited by 0
  4. 2026Global Symmetry and Orthogonal Transformations from Geometrical Moment n-tuplescited by 0
  5. 2026ArchSym: Detecting 3D-Grounded Architectural Symmetries in the Wildcited by 0
  6. 2026ARD-REFSM: Enhancing Reflection Symmetry Detection with Asymmetric Denoising and Rotation Equivariancecited by 0