MEGA-PCC: A Mamba-based Efficient Approach for Joint Geometry and Attribute Point Cloud Compression

TL;DR

MEGA-PCC tackles the challenge of jointly compressing geometry and attributes in point clouds with an end-to-end framework that uses a shared latent representation and dual decoders. It introduces Tri-Mamba-based encoding/decoding and a Mamba-based Entropy Model (MEM) to capture long-range spatial and channel correlations, enabling data-driven bitrate allocation without recoloring. Two-stage training stabilizes optimization and yields strong rate-distortion performance, outperforming G-PCC by substantial margins and remaining competitive with other learning-based methods at lower complexity. The approach offers practical benefits for AI-driven PCC, including reduced pipeline complexity, faster runtimes, and scalable end-to-end optimization for both static and potentially dynamic content.

Abstract

Joint compression of point cloud geometry and attributes is essential for efficient 3D data representation. Existing methods often rely on post-hoc recoloring procedures and manually tuned bitrate allocation between geometry and attribute bitstreams in inference, which hinders end-to-end optimization and increases system complexity. To overcome these limitations, we propose MEGA-PCC, a fully end-to-end, learning-based framework featuring two specialized models for joint compression. The main compression model employs a shared encoder that encodes both geometry and attribute information into a unified latent representation, followed by dual decoders that sequentially reconstruct geometry and then attributes. Complementing this, the Mamba-based Entropy Model (MEM) enhances entropy coding by capturing spatial and channel-wise correlations to improve probability estimation. Both models are built on the Mamba architecture to effectively model long-range dependencies and rich contextual features. By eliminating the need for recoloring and heuristic bitrate tuning, MEGA-PCC enables data-driven bitrate allocation during training and simplifies the overall pipeline. Extensive experiments demonstrate that MEGA-PCC achieves superior rate-distortion performance and runtime efficiency compared to both traditional and learning-based baselines, offering a powerful solution for AI-driven point cloud compression.

Figures (6)

• Figure 1: (a) Existing point cloud joint compression methods rely on recoloring to align geometry and attributes. However, manual bit allocation leads to suboptimal reconstruction. (b) The proposed MEGA-PCC uses a shared latent space and loss-based bit allocation, enabling end-to-end compression without recoloring or post-hoc model matching. (c) Traditional model matching involves an exhaustive search of the optimal pairing (e.g., pairing 5 rate points from PCGCv2 refer8 with ANF-PCAC refer22) and selects top results based on PCQM scores, incurring high computational cost.
• Figure 2: Taxonomy of point cloud compression methods.
• Figure 3: The unified encoder jointly processes geometry and attributes using multi-layer sparse convolutions and Mamba blocks to capture both local and global dependencies, producing a shared latent representation. During decoding, geometry is reconstructed first, and the decoded coordinates guide the attribute decoder, enabling efficient joint compression.
• Figure 4: (a) Serialize point clouds into a sequence while preserving spatial proximity relationships between consecutive elements in the sequence (b) Tri-Mamba, which is used in both encoder and decoder for feature extraction, combines forward, backward, and feature channel scanning to comprehensively understand spatial information and leverage channel-wise information to enrich feature representation.
• Figure 5: Mamba-based Entropy model.