Research map / 3D vision / Face and shape
DiViNeT: 3D Reconstruction from Disparate Views using Neural Template Regularization
Aditya Vora, Akshay Gadi Patil, Hao Zhang
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 |
|---|---|---|---|
| add depth constraint to SDF-only constraint (ablation) | DiViNeT · DTU sparse | Chamfer Distance: 2.70 -> 1.96 | Improved |
| add SDF constraint to depth-only constraint (full pipeline) (ablation) | DiViNeT · DTU sparse | Chamfer Distance: 1.96 -> 1.52 | Improved |
| replace pre-trained TPN (DTU) with TPN re-trained on target data (ablation) | DiViNeT · MobileBrick Bridge | F1: 0.565 -> 0.658 | Improved |
| TPN re-training vs pre-trained on Colosseum (ablation) | DiViNeT · MobileBrick Colosseum | F1: 0.219 -> 0.22 | No clear effect |
| neural template regularization vs MonoSDF | DiViNeT · DTU sparse 3-view | Chamfer Distance (mean): 1.86 -> 1.77 | Improved |
| template regularization vs NeuS in sparse views | DiViNeT · DTU sparse 3-view | Chamfer Distance (mean): 4.78 -> 1.77 | Improved |
| template regularization vs COLMAP in sparse views | DiViNeT · DTU sparse 3-view | Chamfer Distance (mean): 2.56 -> 1.77 | Improved |
| compare to feature-grid Voxurf with dense views | DiViNeT · DTU dense | Chamfer Distance (mean): 0.79 vs 0.76 | Hurt |
| template regularization vs NeuS in dense views | DiViNeT · DTU dense | Chamfer Distance (mean): 0.84 -> 0.79 | Improved |
| compare against GeoNeuS (COLMAP-point regularization) on dense views | DiViNeT · MobileBrick Bridge | F1: 0.74 -> 0.915 | Improved |
| compare against GeoNeuS on Castle | DiViNeT · MobileBrick Castle | F1: 0.457 -> 0.572 | Improved |
| compare against MonoSDF on novel view synthesis | DiViNeT · DTU 3-view NVS | PSNR: 23.64 -> 24.34 | Improved |
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.
- NeuS — NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction (NeurIPS 2021) · better on 2 of 2 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
- SparseRecon: Neural Implicit Surface Reconstruction from Sparse Views with Feature and Depth Consistencies (ICCV 2025 · similarity 0.89)
- SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Views (ECCV 2022 · similarity 0.89)
- SC-NeuS: Consistent Neural Surface Reconstruction from Sparse and Noisy Views (AAAI 2024 · similarity 0.87)
- Decompositional Neural Scene Reconstruction with Generative Diffusion Prior (CVPR 2025 · similarity 0.87)
- FatesGS: Fast and Accurate Sparse-View Surface Reconstruction Using Gaussian Splatting with Depth-Feature Consistency (AAAI 2025 · similarity 0.87)
- UFORecon: Generalizable Sparse-View Surface Reconstruction from Arbitrary and Unfavorable Sets (CVPR 2024 · similarity 0.86)
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.
