Research map / 3D vision / Gaussian splatting
Mango-GS: Enhancing Spatio-Temporal Consistency in Dynamic Scenes Reconstruction using Multi-Frame Node-Guided 4D Gaussian Splatting
Tingxuan Huang, Haowei Zhu, Jun-Hai Yong, Hao Pan, Bin Wang
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 control nodes with spatial k-NN only (no learned affinity) (ablation) | single-frame baseline · HyperNeRF (ablation setting) | PSNR: 25.15 -> 24.52 | Hurt |
| decouple node position and code with learned affinity (ablation) | Mango-GS with nodes · HyperNeRF (ablation setting) | PSNR: 24.52 -> 25.31 | Improved |
| go multi-frame with temporal MLP layers (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR: 25.31 -> 27.30 | Improved |
| add temporal self-attention to multi-frame network (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR: 27.30 -> 27.78 | Improved |
| add top-k hard-frame loss (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR / tLPIPS: PSNR 27.78 -> 28.05; tLPIPS 0.0196 -> 0.0202 | Mixed |
| add motion-aware loss (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR / tLPIPS: PSNR 28.05 -> 28.32; tLPIPS 0.0202 -> 0.0192 | Improved |
| increase time window from 6 to 8 (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR: 28.35 -> 28.24 | Hurt |
| reduce time window from 6 to 2 (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR: 28.35 -> 27.53 | Hurt |
| use K=5 neighbors instead of K=3 (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR: 28.39 -> 27.90 | Hurt |
| increase initial nodes from 2048 to 4096 (ablation) | Mango-GS · HyperNeRF (ablation setting) | PSNR: 28.32 -> 28.25 | No clear effect |
| use only sparse stride sampler instead of 0.7:0.3 mix with interpolation sampler (ablation) | Mango-GS · HyperNeRF-vrig | PSNR: 26.2 -> 25.4 | Hurt |
| raise interpolation sampler ratio to 0.5 (ablation) | Mango-GS · HyperNeRF-vrig | PSNR: 26.2 -> 25.6 | Hurt |
Later papers that compared with this method
1 later accepted paper reports MangoGS as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- MVFusion-GS: Motion-Variance Guided Temporal Attention for High-Quality Dynamic Gaussian Splatting (ECCV 2026 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SC-GS — SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes (CVPR 2024) · better on 3 of 3 shared results
- MotionGS — MotionGS: Exploring Explicit Motion Guidance for Deformable 3D Gaussian Splatting (NeurIPS 2024) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- HAIF-GS: Hierarchical and Induced Flow-Guided Gaussian Splatting for Dynamic Scene (NeurIPS 2025 · similarity 0.92)
- Motion Decoupled 3D Gaussian Splatting for Dynamic Object Representation (AAAI 2025 · similarity 0.92)
- Swift4D: Adaptive divide-and-conquer Gaussian Splatting for compact and efficient reconstruction of dynamic scene (ICLR 2025 · similarity 0.91)
- HiCoM: Hierarchical Coherent Motion for Dynamic Streamable Scenes with 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.91)
- MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes (CVPR 2024 · similarity 0.90)
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.
