Research map / 3D vision / Face and shape
Ladybird: Quasi-Monte Carlo Sampling for Deep Implicit Field Based 3D Reconstruction with Symmetry
Yifan Xu, Tianqi Fan, Yi Yuan, Gurprit Singh
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 |
|---|---|---|---|
| sample with Grid+FPS instead of Grid+Random (ablation) | Ladybird · ShapeNet test | Chamfer distance (x0.001): 10.17 -> 8.43 | Improved |
| sample with Grid+FPS instead of Grid+Random (ablation) | Ladybird · ShapeNet test | EMD (x100): 2.71 -> 2.57 | Improved |
| sample with Grid+FPS instead of Sobol+FPS (ablation) | Ladybird · ShapeNet test | Chamfer distance (x0.001): 11.33 -> 8.43 | Improved |
| sample with Grid+FPS instead of Jitter+FPS (ablation) | Ladybird · ShapeNet test | Chamfer distance (x0.001): 19.88 -> 8.43 | Improved |
| select subset with FPS instead of random from grid points (ablation) | Grid sampler · 2D, 1024 points | star discrepancy (x0.01): 4.51 -> 3.86 | Improved |
| select subset with FPS instead of random (ablation) | Ladybird · SDF validation accuracy, epoch 30 | accuracy: 0.825 -> 0.917 | Improved |
| add Symm(Concat) symmetric feature fusion (ablation) | Ladybird · ShapeNet test | Chamfer distance (x0.001): 8.43 -> 6.33 | Improved |
| add Symm(Concat) symmetric feature fusion (ablation) | Ladybird · ShapeNet test | IoU (%): 61.9 -> 65.6 | Improved |
| add Symm(Near) symmetric feature fusion (ablation) | Ladybird · ShapeNet test | Chamfer distance (x0.001): 8.43 -> 7.66 | Improved |
| fuse symmetric features by averaging (Symm(Avg)) (ablation) | Ladybird · ShapeNet test | Chamfer distance (x0.001): 8.43 -> 19.41 | Hurt |
| use estimated instead of ground truth camera parameters | Ladybird · ShapeNet Core test | Chamfer distance (x0.001): 6.33 -> 8.60 | Hurt |
| raise Marching Cube resolution from 64^3 to 512^3 (ablation) | Ladybird Grid+FPS · ShapeNet 100 objects | Chamfer distance (x0.001): 9.04 -> 8.81 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Pixel2Mesh — Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images (ECCV 2018) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- IPoD: Implicit Field Learning with Point Diffusion for Generalizable 3D Object Reconstruction from Single RGB-D Images (CVPR 2024 · similarity 0.86)
- Differentiable Gradient Sampling for Learning Implicit 3D Scene Reconstructions from a Single Image (ICLR 2022 · similarity 0.85)
- Neural Fields As Learnable Kernels for 3D Reconstruction (CVPR 2022 · similarity 0.85)
- DI-Fusion: Online Implicit 3D Reconstruction With Deep Priors (CVPR 2021 · similarity 0.84)
- SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images (CVPR 2025 · similarity 0.84)
- LISR: Learning Linear 3D Implicit Surface Representation Using Compactly Supported Radial Basis Functions (AAAI 2024 · similarity 0.83)
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.
