Research map / 3D vision / Face and shape
Primitive-based shape abstraction
Represents shapes with learned primitives, parts and implicit grids for retrieval and generation.
99 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Face and shape, keyword cluster “3d shape · shapes · object”
State History Benchmarks Tried Checklist Preprints Who Export
Where it stands
Few recent reports None of the tracked benchmarks has a report from 2025–26.
12 papers in 2023–24 and 12 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic held steady, from 3.3% to 3.3%.
No longer reported by papers here: ModelNet40 (mAP, last in 2019).
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Learning Cross-Domain Neural Networks for Sketch-Based 3D Shape Retrieval (AAAI 2016 · 77 citations)
- 2017 · Shape Completion Using 3D-Encoder-Predictor CNNs and Shape Synthesis (CVPR 2017 · 678 citations)
- 2018 · GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition (CVPR 2018 · 653 citations)
- 2019 · MeshNet: Mesh Neural Network for 3D Shape Representation (AAAI 2019 · 338 citations)
- 2020 · Local Implicit Grid Representations for 3D Scenes (CVPR 2020 · 550 citations)
- 2021 · Deformed Implicit Field: Modeling 3D Shapes With Learned Dense Correspondence (CVPR 2021 · 145 citations)
- 2022 · CAPRI-Net: Learning Compact CAD Shapes With Adaptive Primitive Assembly (CVPR 2022 · 65 citations)
- 2023 · Understanding and Improving Features Learned in Deep Functional Maps (CVPR 2023 · 24 citations)
- 2024 · Diffusion 3D Features (Diff3F): Decorating Untextured Shapes with Distilled Semantic Features (CVPR 2024 · 29 citations)
- 2025 · Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space (CVPR 2025 · 3 citations)
- 2026 · Learning Compact Latent Space for Representing Neural Signed Distance Functions with High-fidelity Geometry Details (AAAI 2026 · 1 citation)
Earliest papers here
- Learning Cross-Domain Neural Networks for Sketch-Based 3D Shape Retrieval (AAAI 2016 · 77 citations)
- Learned Binary Spectral Shape Descriptor for 3D Shape Correspondence (CVPR 2016 · 26 citations)
- 3D Shape Attributes (CVPR 2016 · 24 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
ModelNet40 · mAP (higher is better)
5 papers, 2018–2019 · not reported since 2019
| Year | Best so far | Set by |
|---|---|---|
| 2018 | 85.38 | Emphasizing 3D Properties in Recurrent Multi-View Aggregation for 3D Shape Retrieval (AAAI 2018 · 13 citations) |
| 2019 | 89.23 | View Inter-Prediction GAN: Unsupervised Representation Learning for 3D Shapes by Learning Global Shape Memories to Support Local View Predictions (AAAI 2019 · 134 citations) |
What has been tried
87 papers here report 862 measured changes (a component added, removed or swapped, compared with a baseline): helped 74%, no clear effect 6%, hurt 17%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add ControlNet texturing | SHREC’19 · acc | 17.20 -> 26.41 | Helped | Diffusion 3D Features (Diff3F): Decorating Untextured Shapes with Distilled Semantic Features (CVPR 2024 · 29 citations) |
| distill GLIP predictions into a 3D student rather than transfer boxes directly | ShapeNetPart Chair · mIoU (%) | 60.1 -> 73.5 | Helped | PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation (CVPR 2024 · 18 citations) |
| add the feature adapter | PASCAL3D+ · 30° accuracy | 25.9 -> 75.3 | Helped | Common3D: Self-Supervised Learning of 3D Morphable Models for Common Objects in Neural Feature Space (CVPR 2025 · 3 citations) |
| add diversity constraint | Obstacle (multiple shapes) · δ_mean | 0.12 -> 0.15 | Hurt | Geometry-Informed Neural Networks (ICML 2025 · 2 citations) |
| Replace the VN-SPD constraint with standard VN layers, centering, and normalization | PCN · CD-ℓ1 | 8.59 -> 9.31 | Hurt | Learning Generalizable Shape Completion with SIM(3) Equivariance (NeurIPS 2025 · 0 citations) |
| Add indexation invariance using circular padding and indexation-invariant reconstruction loss | sMNIST · F1-score | 0.88 ± 0.01 -> 0.95 ± 0.00 | Helped | ShapeEmbed: a self-supervised learning framework for 2D contour quantification (NeurIPS 2025 · 0 citations) |
| Use triplet training instead of pairwise training | Homography-Perturbed MNIST · accuracy | 92.52 (± 0.91) -> 97.38 (± 0.23) | Helped | Neural Isometries: Taming Transformations for Equivariant ML (NeurIPS 2024 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 5 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- MPEG-7 CE-Shape-1 Part B · 2 of 5 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 24 papers; from OpenAlex and the papers' affiliations).
- Korea Advanced Institute of Science and Technology · 3 papers
Use it
Download BibTeX of 37 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Single-view 3D reconstruction (96 papers)
- Articulated and non-rigid shape learning (75 papers)
- 3D shape generative modeling (74 papers)
- Optimization-based shape matching (59 papers)
Elsewhere in the topic
- Neural implicit surface reconstruction (131 papers)
- Gaussian and implicit head avatars (115 papers)
- Textured mesh generation (111 papers)
- Face reflectance and relightable rendering (85 papers)
- 3D morphable face model regression (80 papers)
- Full-body avatar reconstruction (78 papers)
- Learned mesh and geometric representations (73 papers)
- Clothed human modeling (60 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).