Research map / 3D vision / Gaussian splatting
GaussianUpdate: Continual 3D Gaussian Splatting Update for Changing Environments
Lin Zeng, Boming Zhao, Jiarui Hu, Xujie Shen, Ziqiang Dang, Hujun Bao, Zhaopeng Cui
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 learnable removal factors (ablation) | GaussianUpdate · WAT dataset | PSNR: 25.87 -> 26.55 | Improved |
| add learnable removal factors (ablation) | GaussianUpdate · WAT dataset | SSIM: 0.834 -> 0.844 | Improved |
| add sparse COLMAP point addition of new Gaussians (ablation) | GaussianUpdate · WAT dataset | PSNR: 25.96 -> 26.55 | Improved |
| add sparse COLMAP point addition of new Gaussians (ablation) | GaussianUpdate · WAT dataset | SSIM: 0.829 -> 0.844 | Improved |
| add layout-invariant mask for appearance stage (ablation) | GaussianUpdate · WAT dataset | PSNR: 26.18 -> 26.55 | Improved |
| add layout-invariant mask for appearance stage (ablation) | GaussianUpdate · WAT dataset | SSIM: 0.839 -> 0.844 | Improved |
| use generative replay instead of memory replay storing all past images | GaussianUpdate · WAT dataset, Breville | PSNR: 30.36 (UB) vs 30.11 | No clear effect |
| use GaussianUpdate instead of CLNeRF | CLNeRF · WAT dataset rendering speed | FPS: 0.75 -> 34 | Improved |
| use GaussianUpdate instead of CLNeRF | CLNeRF · WAT dataset, Kitchen | PSNR: 28.40 -> 28.02 | Hurt |
| use GaussianUpdate instead of CLNeRF | CLNeRF · WAT dataset, Community | PSNR: 22.88 -> 23.88 | Improved |
| use GaussianUpdate instead of 4DGS | 4DGS · Synthetic NeRF, Lego | PSNR: 33.23 -> 35.70 | Improved |
| add generative replay (vs baseline without it) | 3DGS baseline · WAT dataset, Breville | PSNR: 20.66 -> 30.11 | Improved |
Closest papers by meaning
- MaskGaussian: Adaptive 3D Gaussian Representation from Probabilistic Masks (CVPR 2025 · similarity 0.88)
- FewViewGS: Gaussian Splatting with Few View Matching and Multi-stage Training (NeurIPS 2024 · similarity 0.88)
- Decoupling Appearance Variations with 3D Consistent Features in Gaussian Splatting (AAAI 2025 · similarity 0.87)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.87)
- Generative Sparse-View Gaussian Splatting (CVPR 2025 · similarity 0.87)
- Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splatting (NeurIPS 2025 · 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.
