Research map / Point clouds / Point cloud classification
LINR-PCGC: Lossless Implicit Neural Representations for Point Cloud Geometry Compression
Wenjie Huang, Qi Yang, Shuting Xia, He Huang, Yiling Xu, Zhu Li
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 adaptive quantization and model compression (ablation) | LINR-PCGC (CNP only) · avg of 8iVFB, Owlii, MVUB | relative bpp (%): 100.0 -> 91.9 | Improved |
| add scale context extraction (SCE) (ablation) | LINR-PCGC · avg of 8iVFB, Owlii, MVUB | relative bpp (%): 91.9 -> 88.8 | Improved |
| initialize later GoPs from the previous overfitted GoP (ablation) | LINR-PCGC · avg training time | relative time (%): 100 -> 34.7 | Improved |
| also initialize first GoP from a similar sequence (ablation) | LINR-PCGC · avg training time | relative time (%): 34.7 -> 24.0 | Improved |
| add L2 regularization with model compression (ablation) | LINR-PCGC · avg of Owlii, 8iVFB, MVUB | relative bpp: 1.081 -> 1.000 | Improved |
| add model compression without regularization (ablation) | LINR-PCGC · avg of Owlii, 8iVFB, MVUB | relative bpp: 1.090 -> 1.081 | No clear effect |
| add regularization alone without MC (ablation) | LINR-PCGC · avg of Owlii, 8iVFB, MVUB | relative bpp: 1.090 -> 1.089 | No clear effect |
| use Laplace arithmetic coding instead of LZ77 for parameters (ablation) | LINR-PCGC · avg bitstream of parameters (bits) | bits: 268672 -> 251360 | Improved |
| use CNP instead of 8-stage SOPA (ablation) | LINR-PCGC · avg of Owlii and 8iVFB | relative bpp: 1 -> 0.9238 | Improved |
| use CNP instead of 8-stage SOPA (ablation) | LINR-PCGC · avg of Owlii and 8iVFB | peak memory: 11.92 -> 4.54 | Improved |
| more first-GoP training epochs (F 6 to 61) (ablation) | LINR-PCGC · 8iVFB | bpp: 0.616 -> 0.567 | Improved |
| use LINR-PCGC instead of SparsePCGC | SparsePCGC · MVUB | bpp: 0.93 -> 0.806 | Improved |
Closest papers by meaning
- UniPCGC: Towards Practical Point Cloud Geometry Compression via an Efficient Unified Approach (AAAI 2025 · similarity 0.90)
- patchDPCC: A Patchwise Deep Compression Framework for Dynamic Point Clouds (AAAI 2024 · similarity 0.87)
- TopNet: Transformer-Efficient Occupancy Prediction Network for Octree-Structured Point Cloud Geometry Compression (CVPR 2025 · similarity 0.87)
- RENO: Real-Time Neural Compression for 3D LiDAR Point Clouds (CVPR 2025 · similarity 0.86)
- LCM: Locally Constrained Compact Point Cloud Model for Masked Point Modeling (NeurIPS 2024 · similarity 0.85)
- msLPCC: A Multimodal-Driven Scalable Framework for Deep LiDAR Point Cloud Compression (AAAI 2024 · similarity 0.85)
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.
