Research map / 3D vision / Face and shape
Pixel-aligned implicit human and hand reconstruction
Reconstructs humans and hand-held objects with pixel-aligned implicit functions and explicit contact.
39 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2017–2026) · an approach in Face and shape, keyword cluster “surface · mesh · implicit”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
15 papers in 2023–24 and 10 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 4.1% to 2.7%.
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
- 2017 · Self-supervised Learning of Motion Capture (NeurIPS 2017 · 160 citations)
- 2019 · Photometric Mesh Optimization for Video-Aligned 3D Object Reconstruction (CVPR 2019 · 76 citations)
- 2020 · PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization (CVPR 2020 · 800 citations)
- 2021 · Function4D: Real-Time Human Volumetric Capture From Very Sparse Consumer RGBD Sensors (CVPR 2021 · 351 citations)
- 2022 · LISA: Learning Implicit Shape and Appearance of Hands (CVPR 2022 · 69 citations)
- 2023 · gSDF: Geometry-Driven Signed Distance Functions for 3D Hand-Object Reconstruction (CVPR 2023 · 56 citations)
- 2024 · Neural Implicit Representation for Building Digital Twins of Unknown Articulated Objects (CVPR 2024 · 25 citations)
- 2025 · HumanRig: Learning Automatic Rigging for Humanoid Character in a Large Scale Dataset (CVPR 2025 · 4 citations)
- 2026 · ZipMap: Linear-Time Stateful 3D Reconstruction via Test-Time Training (CVPR 2026 · 0 citations)
Earliest papers here
- Self-supervised Learning of Motion Capture (NeurIPS 2017 · 160 citations)
- Photometric Mesh Optimization for Video-Aligned 3D Object Reconstruction (CVPR 2019 · 76 citations)
- PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization (CVPR 2020 · 800 citations)
What has been tried
33 papers here report 315 measured changes (a component added, removed or swapped, compared with a baseline): helped 79%, no clear effect 5%, hurt 15%, mixed 1%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| Add shared motion parameters across the two articulation models | Multi-part objects · Axis Ang | 14.42±0.90 -> 0.26±0.04 | Helped | Neural Implicit Representation for Building Digital Twins of Unknown Articulated Objects (CVPR 2024 · 25 citations) |
| Replace the global-feature token generator and MLP mesh regressor with the proposed mesh regressor | FreiHAND · PA-MPJPE (mm) | 6.9 -> 6.5 | Helped | A Simple Baseline for Efficient Hand Mesh Reconstruction (CVPR 2024 · 21 citations) |
| Add point-feature refinement | ZJU-MOCAP · PSNR | 30.93 -> 31.09 | Helped | HumanNeRF-SE: A Simple yet Effective Approach to Animate HumanNeRF with Diverse Poses (CVPR 2024 · 9 citations) |
| Add HAM to the initialized mesh | Evaluation (P2S and PSNR) · P2S | 1.873 -> 1.457 | Helped | Fine-Grained Multi-View Hand Reconstruction Using Inverse Rendering (AAAI 2024 · 8 citations) |
| Add silhouette loss | ϕ-SfT R3 · symmetric Chamfer distance ×104 | 13.8 -> 12.5 | Helped | Physics-guided Shape-from-Template: Monocular Video Perception through Neural Surrogate Models (CVPR 2024 · 7 citations) |
| add vertex sub-mesh loss | OakInk · F1 | 0.189 -> 0.210 | Helped | Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · 6 citations) |
| Add 3DMM supervision | Self-driven head reconstruction · LPIPS | 0.0686 -> 0.0223 | Helped | Learning Dynamic Tetrahedra for High-Quality Talking Head Synthesis (CVPR 2024 · 6 citations) |
| Add PGSE | HumanRig test · CD-J2J | 0.0110 -> 0.0027 | Helped | HumanRig: Learning Automatic Rigging for Humanoid Character in a Large Scale Dataset (CVPR 2025 · 4 citations) |
| add the thin-shell physics prior | ϕ-SfT dataset · Chamfer distance (×10^4) | 34.25 -> 3.46 | Helped | Thin-Shell-SfT: Fine-Grained Monocular Non-rigid 3D Surface Tracking with Neural Deformation Fields (CVPR 2025 · 2 citations) |
| Add causal consistency loss Lcausal | Triplane interpolation · PSNR | 22.2 -> 23.11 | Helped | LIM: Large Interpolator Model for Dynamic Reconstruction (CVPR 2025 · 1 citation) |
| Add Dynamic Rigidity instead of ARAP | DAVIS · 2D keypoint transfer accuracy | 78.5 -> 80.2 | Helped | Learning Implicit Representation for Reconstructing Articulated Objects (ICLR 2024 · 1 citation) |
| Replace XHand with the enhanced mesh representation | InterHand2.6M reconstruction, Identity 1 · PSNR | 27.77 -> 28.92 | Helped | SRHand: Super-Resolving Hand Images and 3D Shapes via View/Pose-aware Neural Image Representations and Explicit Meshes (NeurIPS 2025 · 0 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 25 papers; from OpenAlex and the papers' affiliations).
- University of Chinese Academy of Sciences · 2 papers
- Inria · 2 papers
Use it
Download BibTeX of 34 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
- Neural implicit surface reconstruction (131 papers)
- Textured mesh generation (111 papers)
- Learned mesh and geometric representations (73 papers)
- Implicit neural representation design (52 papers)
Elsewhere in the topic
- Gaussian and implicit head avatars (115 papers)
- Primitive-based shape abstraction (99 papers)
- Single-view 3D reconstruction (96 papers)
- Face reflectance and relightable rendering (85 papers)
- 3D morphable face model regression (80 papers)
- Full-body avatar reconstruction (78 papers)
- Articulated and non-rigid shape learning (75 papers)
- 3D shape generative modeling (74 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).