Point Cloud Feature Coding for Object Detection over an Error-Prone Cloud-Edge Collaborative System

TL;DR

A lightweight yet effective feature compaction module that compresses the deepest feature among multi-scale representations by removing task-irrelevant regions and applying channel-wise dimensionality reduction to task-relevant areas is designed.

Abstract

Cloud-edge collaboration enhances machine perception by combining the strengths of edge and cloud computing. Edge devices capture raw data (e.g., 3D point clouds) and extract salient features, which are sent to the cloud for deeper analysis and data fusion. However, efficiently and reliably transmitting features between cloud and edge devices remains a challenging problem. We focus on point cloud-based object detection and propose a task-driven point cloud compression and reliable transmission framework based on source and channel coding. To meet the low-latency and low-power requirements of edge devices, we design a lightweight yet effective feature compaction module that compresses the deepest feature among multi-scale representations by removing task-irrelevant regions and applying channel-wise dimensionality reduction to task-relevant areas. Then, a signal-to-noise ratio (SNR)-adaptive channel encoder dynamically encodes the attribute information of the compacted features, while a Low-Density Parity-Check (LDPC) encoder ensures reliable transmission of geometric information. At the cloud side, an SNR-adaptive channel decoder guides the decoding of attribute information, and the LDPC decoder corrects geometry errors. Finally, a feature decompaction module restores the channel-wise dimensionality, and a diffusion-based feature upsampling module reconstructs shallow-layer features, enabling multi-scale feature reconstruction. On the KITTI dataset, our method achieved a 172-fold reduction in feature size with 3D average precision scores of 93.17%, 86.96%, and 77.25% for easy, moderate, and hard objects, respectively, over a 0 dB SNR wireless channel. Our source code will be released on GitHub at: https://github.com/yuanhui0325/T-PCFC.

Figures (13)

• Figure 1: Pipeline of the used object detection network. StVD denotes Stochastic Voxel Discard, NR Conv denotes Noise-Resistant Submanifold Convolution and 3D SpConv denotes 3D Sparse Convolution.
• Figure 2: Overview of the proposed network.
• Figure 3: Channel compaction block.
• Figure 4: Spatial compaction block.
• Figure 5: Dense point mask generation.