Ruixiang Xue薛瑞翔

We introduce a framework towards scalable, finer-grained object-level 3DGS compression. First, a post-training method named RecastGS is proposed to reorganize pretrained 3DGS into a layered representation and progressively distills cumulative submodels to improve rate-distortion efficiency. Leveraging multi-view SAM predictions from user click prompts, Gaussians are further partitioned into user-defined regions of interest (ROI), enabling region-adaptive quality control without retraining. Second, built upon this reorganized region-aware layered hierarchy, a feed-forward 3DGS compression method named LayeredCGS is proposed to compress position using a lightweight point cloud codec and attributes with a layer-wise context model to exploit cross-layer correlations. Extensive experiments show that LayeredCGS achieves 35% BD-Rate gain over the existing feed-forward method FCGS. With progressive distillation in RecastGS enabled, our method further outperforms most per-scene optimization methods. Moreover, the proposed method supports ROI-aware compression and flexible bitstream truncation, achieving up to 2 dB higher ROI PSNR at comparable bitrates compared with the uniform quality allocation baseline while enabling low-latency preview and progressive quality refinement. The code will be released at https://github.com/RuixiangXue/ScalableGSC.

A universal multiscale conditional coding framework, Unicorn, is proposed to compress the geometry and attribute of any given point cloud. Geometry compression is addressed in Part I of this paper, while attribute compression is discussed in Part II. We construct the multiscale sparse tensors of each voxelized point cloud frame and properly leverage lower-scale priors in the current and (previously processed) temporal reference frames to improve the conditional probability approximation or content-aware predictive reconstruction of geometry occupancy in compression. Unicorn is a versatile, learning-based solution capable of compressing static and dynamic point clouds with diverse source characteristics in both lossy and lossless modes. Following the same evaluation criteria, Unicorn significantly outperforms standard-compliant approaches like MPEG G-PCC, V-PCC, and other learning-based solutions, yielding state-of-the-art compression efficiency while presenting affordable complexity for practical implementations.

This paper proposes the NeRI, an implicit neural representation (INR) based LiDAR point cloud compressor. In NeRI, we first transform a sequence of 3D LiDAR frames into a 2D range image sequence through range image projection over time. Then, we employ a neural network conditioned on the temporal frame index and associated LiDAR sensor pose to fit input range images as closely as possible. The optimized network parameters, which implicitly represent the input LiDAR data, are later lossily compressed. NeRI decoder is then initialized using decoded parameters to generate range images for reconstructing the 3D LiDAR sequence accordingly. Extensive experimental results demonstrate the significant superiority of NeRI regarding the compression efficiency and decoding speed compared to state-of-the-art 2D and 3D compressors for LiDAR point cloud.

A universal multiscale conditional coding framework, Unicorn, is proposed to code the geometry and attribute of any given point cloud. Attribute compression is discussed in Part II of this paper, while geometry compression is given in Part I of this paper. We first construct the multiscale sparse tensors of each voxelized point cloud attribute frame. Since attribute components exhibit very different intrinsic characteristics from the geometry element, e.g., 8-bit RGB color versus 1-bit occupancy, we process the attribute residual between lower-scale reconstruction and current-scale data. Similarly, we leverage spatially lower-scale priors in the current frame and (previously processed) temporal reference frame to improve the probability estimation of attribute intensity through conditional residual prediction in lossless mode or enhance the attribute reconstruction through progressive residual refinement in lossy mode for better performance. The proposed Unicorn is a versatile, learning-based solution capable of compressing a great variety of static and dynamic point clouds in both lossy and lossless modes. Following the same evaluation criteria, Unicorn significantly outperforms standard-compliant approaches like MPEG G-PCC, V-PCC, and other learning-based solutions, yielding state-of-the-art compression efficiency with affordable encoding/decoding runtime.

The lossy Geometry-based Point Cloud Compression (G-PCC) inevitably impairs the geometry information of point clouds, which deteriorates the quality of experience (QoE) in reconstruction and/or misleads decisions in tasks such as classification. To tackle it, this work proposes GRNet for the geometry restoration of G-PCC compressed large-scale point clouds. By analyzing the content characteristics of original and G-PCC compressed point clouds, we attribute the G-PCC distortion to two key factors: point vanishing and point displacement. Visible impairments on a point cloud are usually dominated by an individual factor or superimposed by both factors, which are determined by the density of the original point cloud. To this end, we employ two different models for coordinate reconstruction, termed Coordinate Expansion and Coordinate Refinement, to attack the point vanishing and displacement, respectively. In addition, 4-byte auxiliary density information is signaled in the bitstream to assist the selection of Coordinate Expansion, Coordinate Refinement, or their combination. Before being fed into the coordinate reconstruction module, the G-PCC compressed point cloud is first processed by a Feature Analysis Module for multiscale information fusion, in which kNN-based Transformer is leveraged at each scale to adaptively characterize neighborhood geometric dynamics for effective restoration. Following the common test conditions recommended in the MPEG standardization committee, GRNet significantly improves the G-PCC anchor and remarkably outperforms state-of-the-art methods on a great variety of point clouds (e.g., solid, dense, and sparse samples) both quantitatively and qualitatively. Meanwhile, GRNet runs fairly fast and uses a smaller-size model when compared with existing learning-based approaches, making it attractive to industry practitioners.