Research map / Point clouds / Point cloud classification
AdaDPCC: Adaptive Rate Control and Rate-Distortion-Complexity Optimization for Dynamic Point Cloud Compression
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes AdaDPCC, a slimmable dynamic point cloud codec with multiple coding routes, coarse-to-fine motion estimation/compensation, and a sliding-window rate control module that picks routes to hit target bitrates.
Problem
Dynamic point cloud compression lacks flexible variable-bitrate and computational-complexity control, suffers from sparse, KNN-limited inter-frame prediction that is slow, and has imprecise rate control.
Approach
A slimmable VAE-based codec whose subnets (routes) are jointly trained with cumulative RD losses so one model supports rate-distortion-complexity trade-offs; inter prediction uses sparse convolutions in a coarse-to-fine motion scheme instead of KNN; a rate estimator plus sliding-window bit allocation picks the route per frame.
Key results
- Average BD-Rate reduction of 5.81% and BD-PSNR gain of 0.42 dB versus patchDPCC on 8iVFB (D1).
- Average bitrate error of 0.40% for rate control across targets 0.15-0.29 bpp.
- Average coding time reduced by up to 44.6% compared to D-DPCC (Route 0 vs D-DPCC).
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use AdaDPCC instead of patchDPCC | AdaDPCC · 8iVFB | D1 BD-Rate (%): -5.810% | Helped |
| Use AdaDPCC instead of patchDPCC | AdaDPCC · 8iVFB | D1 BD-PSNR (dB): +0.423 dB | Helped |
| Use AdaDPCC instead of D-DPCC | AdaDPCC Route 3 · 8iVFB | coding time: 8.2% faster | Helped |
| Select Route 0 instead of Route 3 | AdaDPCC · 8iVFB | coding time: 39.4% reduction | Helped |
| Use AdaDPCC coarse-to-fine inter-frame prediction instead of 3DAWI (ablation) | AdaDPCC · Soldier | encoding time: 8.8% of the time needed by 3DAWI | Helped |
| Add coarse-to-fine inter-frame prediction instead of 3DAWI (ablation) | AdaDPCC · 8iVFB | BD-Rate reduction: 3.4% | Helped |
| Add hyper-prior context (ablation) | AdaDPCC · 8iVFB | BD-Rate reduction: 2.9% | Helped |
Limitations
- Results for patchDPCC are taken from the original paper because its source code is unavailable.
- Only geometry (D1/D2) evaluated on 8iVFB; no other datasets reported in the text.
Experimental setup
Datasets: Owlii MPEG 8iVFB
Benchmarks: MPEG 8iVFB
Metrics: BD-Rate BD-PSNR D1 PSNR D2 PSNR bitrate error coding time
Baselines: D-DPCC patchDPCC PCGCv2 V-PCC
Builds on: D-DPCC PCGCv2 3DAWI slimmable networks
Closest papers by meaning
- patchDPCC: A Patchwise Deep Compression Framework for Dynamic Point Clouds (AAAI 2024 · similarity 0.85)
- Fast Inter-frame Motion Prediction for Compressed Dynamic Point Cloud Attribute Enhancement (AAAI 2024 · similarity 0.85)
- UniPCGC: Towards Practical Point Cloud Geometry Compression via an Efficient Unified Approach (AAAI 2025 · similarity 0.85)
- Task-Aware Encoder Control for Deep Video Compression (CVPR 2024 · similarity 0.83)
- 4DGC: Rate-Aware 4D Gaussian Compression for Efficient Streamable Free-Viewpoint Video (CVPR 2025 · similarity 0.83)
- 4DGCPro: Efficient Hierarchical 4D Gaussian Compression for Progressive Volumetric Video Streaming (NeurIPS 2025 · similarity 0.82)
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.