Research map / Point clouds / Point cloud classification
NPBG++: Accelerating Neural Point-Based Graphics
Check 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 |
|---|---|---|---|
| Add view-dependent MLP aggregation with m=6 basis functions instead of average aggregation (ablation) | NPBG++ · ScanNet holdout scenes 0, 43, 45 | PSNR: 21.41 -> 23.11 | Helped |
| Use spherical harmonics (m=4) instead of average aggregation (ablation) | NPBG++ · ScanNet holdout scenes 0, 43, 45 | PSNR: 21.41 -> 22.19 | Helped |
| Use MLP aggregation with m=3 instead of average aggregation (ablation) | NPBG++ · ScanNet holdout scenes 0, 43, 45 | PSNR: 21.41 -> 22.97 | Helped |
| Use MLP aggregation with m=6 instead of m=3 (ablation) | NPBG++ · ScanNet holdout scenes 0, 43, 45 | PSNR: 22.97 -> 23.11 | Helped |
| Add both input and output image alignment rather than using neither (ablation) | NPBG++ · DTU holdout scenes 110, 114, 118 | PSNR: 22.53 -> 23.23 | Helped |
| Use input image alignment only rather than neither alignment (ablation) | NPBG++ · DTU holdout scenes 110, 114, 118 | PSNR: 22.53 -> 21.87 | Hurt |
| Use output image alignment only rather than neither alignment (ablation) | NPBG++ · DTU holdout scenes 110, 114, 118 | PSNR: 22.53 -> 22.47 | No clear effect |
| Add output image alignment (ablation) | NPBG · DTU holdout scenes 110, 114, 118 | PSNR: 26.00 -> 26.16 | Helped |
| Finetune coefficients and refiner after feedforward scene fitting | NPBG++ · H3DS holdout | PSNR: 21.80 -> 24.91 | Helped |
| Finetune coefficients and refiner after feedforward scene fitting | NPBG++ · DTU holdout | PSNR: 23.23 -> 26.08 | Helped |
| Finetune coefficients and refiner after feedforward scene fitting | NPBG++ · ScanNet holdout | PSNR: 23.11 -> 25.27 | Helped |
| Finetune coefficients and refiner after feedforward scene fitting | NPBG++ · NeRF-Synthetic holdout | PSNR: 26.06 -> 28.67 | Helped |
Closest papers by meaning
- HashPoint: Accelerated Point Searching and Sampling for Neural Rendering (CVPR 2024 · similarity 0.87)
- GoLF-NRT: Integrating Global Context and Local Geometry for Few-Shot View Synthesis (CVPR 2025 · similarity 0.87)
- NeRF-VPT: Learning Novel View Representations with Neural Radiance Fields via View Prompt Tuning (AAAI 2024 · similarity 0.86)
- NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis (NeurIPS 2026 · similarity 0.86)
- Efficient-3Dim: Learning a Generalizable Single-image Novel-view Synthesizer in One Day (ICLR 2024 · similarity 0.86)
- UMAMI: Unifying Masked Autoregressive Models and Deterministic Rendering for View Synthesis (NeurIPS 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.