Research map / 3D vision / Face and shape
NU-MCC: Multiview Compressive Coding with Neighborhood Decoder and Repulsive UDF
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use the Neighborhood decoder with coarse anchor features instead of MCC's decoder (ablation) | NU-MCC · CO3D-v2 validation | inference speed: 15× inference speedup | Helped |
| add fine input features to the feature aggregation (ablation) | NU-MCC Neighborhood decoder · CO3D-v2 validation | color accuracy | Helped |
| add fine input features to the feature aggregation (ablation) | NU-MCC Neighborhood decoder · CO3D-v2 validation | geometry accuracy | Helped |
| replace occupancy with standard UDF (ablation) | NU-MCC · CO3D-v2 validation | geometry and color quality | Mixed |
| add repulsive forces between nearby points to standard UDF (ablation) | NU-MCC · CO3D-v2 validation | surface reconstruction quality | Helped |
| use NU-MCC instead of MCC | single-view scene reconstruction · Hypersim validation | F1: 31.25±1.1 -> 49.74±1.7; improves by 59% | Helped |
| use NU-MCC instead of MCC | single-view scene reconstruction · Hypersim validation | L1-RGB: 0.807±0.020 -> 0.605±0.020; reduces by 25% | Helped |
| increase the fine-feature resolution at test time from 112 × 112 to 224 × 224 (ablation) | NU-MCC · CO3D-v2 high-resolution reconstruction | reconstruction detail: 112 × 112 -> 224 × 224 | Helped |
| increase the number of aggregated features at test time (ablation) | NU-MCC · noisy-depth input | reconstruction smoothness: 12 coarse and fine features | Helped |
Closest papers by meaning
- 2S-UDF: A Novel Two-stage UDF Learning Method for Robust Non-watertight Model Reconstruction from Multi-view Images (CVPR 2024 · similarity 0.88)
- Sparse3D: Distilling Multiview-Consistent Diffusion for Object Reconstruction from Sparse Views (AAAI 2024 · similarity 0.87)
- Slice3D: Multi-Slice Occlusion-Revealing Single View 3D Reconstruction (CVPR 2024 · similarity 0.86)
- MV-DUSt3R+: Single-Stage Scene Reconstruction from Sparse Views In 2 Seconds (CVPR 2025 · similarity 0.86)
- Decompositional Neural Scene Reconstruction with Generative Diffusion Prior (CVPR 2025 · similarity 0.86)
- CAT-3DGS: A Context-Adaptive Triplane Approach to Rate-Distortion-Optimized 3DGS Compression (ICLR 2025 · 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.