Research map / 3D vision / Gaussian splatting
FPGS: Feed-Forward Semantic-aware Photorealistic Style Transfer of Large-Scale Gaussian Splatting
GeonU Kim, Kim Youwang, Lee Hyoseok, Tae-Hyun Oh
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 |
|---|---|---|---|
| use FPGS instead of FPRF | FPRF · LLFF | multi-view error (short): 0.0231 -> 0.0214 | Improved |
| use FPGS instead of StyleGaussian | StyleGaussian · LLFF | multi-view error (long): 0.0440 -> 0.0349 | Improved |
| use FPGS instead of FPRF | FPRF · San Francisco Mission Bay | multi-view error (long): 0.0563 -> 0.0502 | Improved |
| use FPGS instead of FPRF | FPRF · LLFF | total training time (min): 61 -> 24 | Improved |
| use FPGS instead of FPRF | FPRF · LLFF | render time (sec): 3.812 -> 0.004 | Improved |
| use FPGS instead of UPST-NeRF | UPST-NeRF · LLFF user study | PST quality rating: 3.279 -> 3.936 | Improved |
| add semantic matching with semantic-weighted local AdaIN (ablation) | FPGS · LLFF | perceptual style similarity | Improved |
| add iterative style transfer (ablation) | FPGS · LLFF | local style transfer quality | Improved |
| use DINO instead of Stable Diffusion v1-5 or DINOv2 as semantic encoder (ablation) | FPGS · LLFF | stylization quality (qualitative) | Improved |
Closest papers by meaning
- FPRF: Feed-Forward Photorealistic Style Transfer of Large-Scale 3D Neural Radiance Fields (AAAI 2024 · similarity 0.94)
- ReGS: Reference-based Controllable Scene Stylization with Gaussian Splatting (NeurIPS 2024 · similarity 0.90)
- Multi-StyleGS: Stylized Gaussian Splatting with Multiple Styles (AAAI 2025 · similarity 0.90)
- SGSST: Scaling Gaussian Splatting Style Transfer (CVPR 2025 · similarity 0.89)
- CLIPGaussian: Universal and Multimodal Style Transfer Based on Gaussian Splatting (NeurIPS 2025 · similarity 0.89)
- Styl3R: Instant 3D Stylized Reconstruction for Arbitrary Scenes and Styles (NeurIPS 2025 · similarity 0.88)
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.
