Research map / Point clouds / Point cloud classification
Point2Pix: Photo-Realistic Point Cloud Rendering via Neural Radiance Fields
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 Point2Pix rather than the strongest listed point-renderer baseline | Point2Pix · ScanNet | PSNR: 17.53 -> 18.47 | Helped |
| use Point2Pix rather than the strongest listed point-renderer baseline | Point2Pix · ARKitScenes | PSNR: 17.61 -> 18.84 | Helped |
| use Point2Pix rather than Point-NeRF after 20 minutes of finetuning | Point2Pix · ScanNet | PSNR: 22.55 -> 23.02 | Helped |
| replace point-guided sampling with uniform sampling (ablation) | Point2Pix · ScanNet | PSNR: 18.47 -> 17.96 | Hurt |
| use point-guided sampling instead of uniform sampling (ablation) | Point2Pix · ScanNet | Render time (seconds): 3.13 -> 0.92 | Helped |
| use point-guided sampling instead of uniform sampling (ablation) | Point2Pix · ScanNet | Train memory (GB): 8.54 -> 6.68 | Helped |
| increase radiance-field scales from one to four at 80 × 60 rays (ablation) | Point2Pix · ScanNet | PSNR: 18.05 -> 18.47 | Helped |
| replace MinkUnet14A with PointNet++ (ablation) | Point2Pix · ScanNet | PSNR: 18.47 -> 16.53 | Hurt |
| replace U-Net with PixelShuffle while retaining concatenation (ablation) | Point2Pix · ScanNet | PSNR: 18.04 -> 18.13 | Helped |
| replace concatenation with SPADE (LayerNorm) conditional fusion (ablation) | PixelShuffle decoder · ScanNet | PSNR: 18.13 -> 18.47 | Helped |
| add point cloud loss with λ_pc = 0.1 (ablation) | Point2Pix · ScanNet | PSNR: 18.23 -> 18.47 | Helped |
| increase point cloud loss weight from 0.1 to 1.0 (ablation) | Point2Pix · ScanNet | PSNR: 18.47 -> 18.30 | Hurt |
Closest papers by meaning
- Sparse Point Cloud Patches Rendering via Splitting 2D Gaussians (CVPR 2025 · similarity 0.90)
- Depth-Guided Robust and Fast Point Cloud Fusion NeRF for Sparse Input Views (AAAI 2024 · similarity 0.89)
- Point2Real: Bridging the Gap between Point Cloud and Realistic Image for Open-World 3D Recognition (AAAI 2024 · similarity 0.87)
- HashPoint: Accelerated Point Searching and Sampling for Neural Rendering (CVPR 2024 · similarity 0.87)
- Neural Point Cloud Diffusion for Disentangled 3D Shape and Appearance Generation (CVPR 2024 · similarity 0.87)
- NexusGS: Sparse View Synthesis with Epipolar Depth Priors in 3D Gaussian Splatting (CVPR 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.