Research map / 3D vision / Gaussian splatting
HAC++: Towards 100X Compression of 3D Gaussian Splatting
Yihang Chen, Qianyi Wu, Weiyao Lin, Mehrtash Tafazzoli Harandi, Jianfei Cai
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 hash-grid context (HAC information) (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +63.3% | Improved |
| add intra-anchor context model (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +14.7% | Improved |
| fuse contexts with GMM instead of concatenation (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +5.7% | Improved |
| apply intra-anchor context to scaling (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: -0.1% | No clear effect |
| apply intra-anchor context to offset (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +0.9% | No clear effect |
| add adaptive offset masking (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +31.4% | Improved |
| add anchor-level mask in entropy loss (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +9.3% | Improved |
| use mask-aware rate instead of extra mask loss term (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +9.6% | Improved |
| add GPCC for anchor location coding (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: +14.0% | Improved |
| disable adaptive quantization module (ablation) | HAC++ · Mip-NeRF360 ablation | fidelity | Improved |
| train 60k instead of 30k iterations (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: -32.9% | Improved |
| train 40k instead of 30k iterations (ablation) | HAC++ · Mip-NeRF360 ablation | BD-rate: -17.5% | Improved |
Closest papers by meaning
- ContextGS : Compact 3D Gaussian Splatting with Anchor Level Context Model (NeurIPS 2024 · similarity 0.92)
- CodecSplat: Ultra-Compact Latent Coding for Feed-Forward 3D Gaussian Splatting (NeurIPS 2026 · similarity 0.91)
- Fast Feedforward 3D Gaussian Splatting Compression (ICLR 2025 · similarity 0.91)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.90)
- Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis (CVPR 2024 · similarity 0.90)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · 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.
