Research map / 3D vision / Gaussian splatting
Complex-Valued 2D Gaussian Representation for Computer-Generated Holography
Yicheng Zhan, Xiangjun Gao, Long Quan, Kaan Akşit
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 structural guidance in POH conversion (vs independently optimized Random POH) (ablation) | Complex-valued 2D Gaussians · Random POH | PSNR: 19.1 -> 30.6 | Improved |
| add structural guidance in POH conversion (ablation) | Complex-valued 2D Gaussians · Random POH | LPIPS: 0.52 -> 0.22 | Improved |
| use unified complex-valued Gaussians instead of naive paired real Gaussians (ablation) | Complex-valued 2D Gaussians · Gaussian definition | PSNR: 25.5 -> 31.8 | Improved |
| use unified complex-valued Gaussians instead of naive paired real Gaussians (ablation) | Complex-valued 2D Gaussians · Gaussian definition | Params (M): 1.2 -> 0.8 | Improved |
| use unified complex-valued Gaussians instead of naive paired real Gaussians (ablation) | Complex-valued 2D Gaussians · Gaussian definition | Render (ms): 20.1 -> 2.13 | Improved |
| keep amplitude decomposed into opacity alpha and color c (ablation) | Complex-valued 2D Gaussians · 50 DIV2K images | PSNR: 29.5 -> 30.7 | Improved |
| keep amplitude decomposed into opacity alpha and color c (ablation) | Complex-valued 2D Gaussians · 50 DIV2K images | LPIPS: 0.40 -> 0.33 | Improved |
| reduce parameters 2:1 vs dense per-pixel (ablation) | Complex-valued 2D Gaussians · Parameter reduction ratio | PSNR: 32.3 -> 31.9 | No clear effect |
| reduce parameters 5:1 vs dense per-pixel (ablation) | Complex-valued 2D Gaussians · Parameter reduction ratio | PSNR: 32.3 -> 30.7 | Hurt |
| reduce parameters 10:1 vs dense per-pixel (ablation) | Complex-valued 2D Gaussians · Parameter reduction ratio | PSNR: 32.3 -> 29.4 | Hurt |
| reduce parameters 10:1 vs dense per-pixel (ablation) | Complex-valued 2D Gaussians · Parameter reduction ratio | Render (ms): 4.02 -> 1.72 | Improved |
| use CUDA kernel instead of PyTorch implementation | Complex-valued 2D Gaussians · runtime | VRAM: 28.9% to 35.6% memory savings | Improved |
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 2 of 2 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
- Compact 3D Gaussian Representation for Radiance Field (CVPR 2024 · similarity 0.86)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.85)
- pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction (CVPR 2024 · similarity 0.84)
- Textured Gaussians for Enhanced 3D Scene Appearance Modeling (CVPR 2025 · similarity 0.84)
- Large Images Are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting (AAAI 2025 · similarity 0.84)
- Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesis (NeurIPS 2024 · similarity 0.84)
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.
