DetVPCC: RoI-based Point Cloud Sequence Compression for 3D Object Detection

2025-02-07

AI Summary

Existing VPCC methods struggle to balance compression efficiency and 3D detection accuracy, primarily due to the lack of discriminative modeling of point cloud region importance. To address this, we propose a Region-of-Interest (ROI)-aware quality-layered coding framework that extends the MPEG VPCC standard for detection-oriented point cloud compression. Our method is the first to integrate ROI awareness directly into VPCC: it employs a lightweight ROI detector to guide spatially non-uniform quantization and explicitly incorporates detection task sensitivity into the encoding pipeline. Leveraging nuScenes data-driven ROI importance estimation and adaptive quality allocation, our approach achieves a 12.6% improvement in BEV mAP and a 4.2% gain in overall 3D detection mAP at identical bitrates—demonstrating significant co-optimization of compression efficiency and detection performance.

Technology Category

• Computer Vision: 3D Computer Vision
• Data Mining & Knowledge Management: Data Compression
• Machine Learning: Learning on the Edge & Model Compression

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

While MPEG-standardized video-based point cloud compression (VPCC) achieves high compression efficiency for human perception, it struggles with a poor trade-off between bitrate savings and detection accuracy when supporting 3D object detectors. This limitation stems from VPCC's inability to prioritize regions of different importance within point clouds. To address this issue, we propose DetVPCC, a novel method integrating region-of-interest (RoI) encoding with VPCC for efficient point cloud sequence compression while preserving the 3D object detection accuracy. Specifically, we augment VPCC to support RoI-based compression by assigning spatially non-uniform quality levels. Then, we introduce a lightweight RoI detector to identify crucial regions that potentially contain objects. Experiments on the nuScenes dataset demonstrate that our approach significantly improves the detection accuracy. The code and demo video are available in supplementary materials.

Problem

• Improves 3D object detection accuracy
• Optimizes point cloud compression efficiency
• Focuses on regions of interest encoding

Innovation

• RoI-based point cloud compression
• Spatially non-uniform quality levels
• Lightweight RoI detector integration