Research map / Point clouds / Point cloud classification
Learning To Stylize Novel Views
Hsin-Ping Huang, Hung-Yu Tseng, Saurabh Saini, Maneesh Singh, Ming-Hsuan Yang
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 point cloud aggregation modules (ablation) | point cloud stylization · Tanks and Temples short-range | warping error: 0.168 -> 0.165 | Improved |
| add point cloud aggregation modules (ablation) | point cloud stylization · Tanks and Temples long-range | warping error: 0.434 -> 0.428 | Improved |
| add point cloud aggregation modules (ablation) | point cloud stylization · Tanks and Temples M60 long-range | warping error: 0.409 -> 0.434 | Hurt |
| use point cloud stylization instead of SVS->LST image stylization after view synthesis | 3D scene stylization · Tanks and Temples long-range | warping error: 0.489 -> 0.431 | Improved |
| use point cloud stylization instead of SVS->Compound video stylization (long-range) | 3D scene stylization · Tanks and Temples long-range | warping error: 0.449 -> 0.431 | Improved |
| use point cloud stylization instead of SVS->Compound video stylization (short-range) | 3D scene stylization · Tanks and Temples short-range | warping error: 0.172 -> 0.170 | No clear effect |
| use point cloud stylization instead of SVS->FMVST | 3D scene stylization · Tanks and Temples short-range | warping error: 0.354 -> 0.170 | Improved |
| use point cloud stylization instead of SVS->MCC | 3D scene stylization · Tanks and Temples long-range | warping error: 0.548 -> 0.431 | Improved |
| use point cloud stylization instead of SVS->TPFR | 3D scene stylization · Tanks and Temples short-range | warping error: 0.264 -> 0.170 | Improved |
| use SVS instead of NeRF++ as the view synthesis stage before LST | image stylization baseline · Tanks and Temples short-range | warping error: 0.231 -> 0.206 | Improved |
| use video stylization (Compound) after view synthesis instead of image stylization (LST) | view synthesis then stylization · Tanks and Temples long-range | warping error: 0.489 -> 0.449 | Improved |
| use deeper VGG feature (relu4_1) instead of relu3_1 for the point cloud (ablation) | point cloud features · stylization level (qualitative) | visual style | Mixed |
Closest papers by meaning
- Styl3R: Instant 3D Stylized Reconstruction for Arbitrary Scenes and Styles (NeurIPS 2025 · similarity 0.91)
- PNeSM: Arbitrary 3D Scene Stylization via Prompt-Based Neural Style Mapping (AAAI 2024 · similarity 0.89)
- Morpheus: Text-Driven 3D Gaussian Splat Shape and Color Stylization (CVPR 2025 · similarity 0.88)
- ReGS: Reference-based Controllable Scene Stylization with Gaussian Splatting (NeurIPS 2024 · similarity 0.87)
- S-DyRF: Reference-Based Stylized Radiance Fields for Dynamic Scenes (CVPR 2024 · similarity 0.87)
- FPRF: Feed-Forward Photorealistic Style Transfer of Large-Scale 3D Neural Radiance Fields (AAAI 2024 · similarity 0.87)
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.
