Research map / 3D vision / Gaussian splatting
GaussianImage++: Boosted Image Representation and Compression with 2D Gaussian Splatting
Tiantian Li, Xinjie Zhang, Xingtong Ge, Tongda Xu, Dailan He, Jun Zhang, Yan 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 growing and pruning densification (ablation) | GaussianImage (5k GS) · Kodak | PSNR: 29.85 -> 31.75 | Improved |
| add growing only (ablation) | no densification · Kodak (5k GS) | PSNR: 29.85 -> 31.71 | Improved |
| add pruning only (ablation) | no densification · Kodak (5k GS) | PSNR: 29.85 -> 29.89 | No clear effect |
| add densification (ablation) | GaussianImage++ · Kodak (5k GS) | encode time (s): 102.78 -> 118.52 | Hurt |
| use LSQ+ instead of FP16 for position (ablation) | GaussianImage++ quantization · Kodak | BD-Rate (%): 25.11 -> 0 | Improved |
| use LSQ+ instead of RVQ for color (ablation) | GaussianImage++ quantization · Kodak | BD-Rate (%): 147.24 -> 0 | Improved |
| use bit-depth 12/10/6 instead of 12/8/8 (ablation) | GaussianImage++ · Kodak | BD-Rate (%): 29.42 -> 0 | Improved |
| use bit-depth 12/10/6 instead of 16/10/6 (ablation) | GaussianImage++ · Kodak | BD-Rate (%): 9.68 -> 0 | Improved |
| use smaller pruning/growing steps tp=100, tg=5000 instead of 500/10000 (ablation) | GaussianImage++ densification · Kodak | PSNR: 31.56 -> 31.83 | Improved |
| replace GaussianImage with GaussianImage++ | GaussianImage · Kodak (10k GS) | PSNR: 32.48 -> 35.41 | Improved |
| replace GaussianImage with GaussianImage++ | GaussianImage · DIV2K (10k GS) | PSNR: 26.54 -> 28.14 | Improved |
| replace GaussianImage with GaussianImage++ | GaussianImage · Kodak (5k GS) | MS-SSIM: 0.962 -> 0.960 | No clear effect |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GaussianImage — GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian Splatting (ECCV 2024) · better on 7 of 8 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
- GaussianSR: High Fidelity 2D Gaussian Splatting for Arbitrary-Scale Image Super-Resolution (AAAI 2025 · similarity 0.91)
- Large Images Are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting (AAAI 2025 · similarity 0.90)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.90)
- Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis (CVPR 2024 · similarity 0.90)
- Compact 3D Gaussian Representation for Radiance Field (CVPR 2024 · similarity 0.89)
- Locality-aware Gaussian Compression for Fast and High-quality Rendering (ICLR 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.
