# Towards Point Cloud Compression for Machine Perception: A Simple and Strong Baseline by Learning the Octree Depth Level Predictor

arXiv: https://arxiv.org/abs/2406.00791 · HTML: https://mlsanity.com/paper/2406.00791/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 25 of its 36 references that are in the map.
- First posted on arXiv 2 Jun 2024.

## Builds on (most cited first, 25 of 25)

- [PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space](https://mlsanity.com/paper/1706.02413/) (2017)
- [PointRCNN: 3D Object Proposal Generation and Detection from Point Cloud](https://mlsanity.com/paper/1812.04244/) (2018)
- [End-to-end Optimized Image Compression](https://mlsanity.com/paper/1611.01704/) (2016)
- [Deep Hough Voting for 3D Object Detection in Point Clouds](https://mlsanity.com/paper/1904.09664/) (2019)
- [Multiscale Point Cloud Geometry Compression](https://mlsanity.com/paper/2011.03799/) (2020)
- [OctSqueeze: Octree-Structured Entropy Model for LiDAR Compression](https://mlsanity.com/paper/2005.07178/) (2020)
- [Scalable Image Coding for Humans and Machines](https://mlsanity.com/paper/2107.08373/) (2021)
- [VoxelContext-Net: An Octree based Framework for Point Cloud Compression](https://mlsanity.com/paper/2105.02158/) (2021)
- [Learned Point Cloud Geometry Compression](https://mlsanity.com/paper/1909.12037/) (2019)
- [Improving Deep Video Compression by Resolution-adaptive Flow Coding](https://mlsanity.com/paper/2009.05982/) (2020)
- [Variable-Rate Deep Image Compression through Spatially-Adaptive Feature Transform](https://mlsanity.com/paper/2108.09551/) (2021)
- [OctAttention: Octree-Based Large-Scale Contexts Model for Point Cloud Compression](https://mlsanity.com/paper/2202.06028/) (2022)
- [Towards Image Understanding from Deep Compression without Decoding](https://mlsanity.com/paper/1803.06131/) (2018)
- [Image coding for machines: an end-to-end learned approach](https://mlsanity.com/paper/2108.09993/) (2021)
- [MuSCLE: Multi Sweep Compression of LiDAR using Deep Entropy Models](https://mlsanity.com/paper/2011.07590/) (2020)
- [Density-preserving Deep Point Cloud Compression](https://mlsanity.com/paper/2204.12684/) (2022)
- [Towards Analysis-friendly Face Representation with Scalable Feature and Texture Compression](https://mlsanity.com/paper/2004.10043/) (2020)
- [Towards End-to-End Image Compression and Analysis with Transformers](https://mlsanity.com/paper/2112.09300/) (2021)
- [CRA5: Extreme Compression of ERA5 for Portable Global Climate and Weather Research via an Efficient Variational Transformer](https://mlsanity.com/paper/2405.03376/) (2024)
- [RIDDLE: Lidar Data Compression with Range Image Deep Delta Encoding](https://mlsanity.com/paper/2206.01738/) (2022)
- [Improving RGB-D Point Cloud Registration by Learning Multi-scale Local Linear Transformation](https://mlsanity.com/paper/2208.14893/) (2022)
- [Group-aware Parameter-efficient Updating for Content-Adaptive Neural Video Compression](https://mlsanity.com/paper/2405.04274/) (2024)
- [Point Cloud Compression with Sibling Context and Surface Priors](https://mlsanity.com/paper/2205.00760/) (2022)
- [Improving Facial Attribute Recognition by Group and Graph Learning](https://mlsanity.com/paper/2105.13825/) (2021)
- [Neural Vector Fields: Generalizing Distance Vector Fields by Codebooks and Zero-Curl Regularization](https://mlsanity.com/paper/2309.01512/) (2023)

## Built on by (most cited first, 4 of 4)

- [HEMGS: A Hybrid Entropy Model for 3D Gaussian Splatting Data Compression](https://mlsanity.com/paper/2411.18473/) (2024)
- [3D Gaussian Splatting Data Compression with Mixture of Priors](https://mlsanity.com/paper/2505.03310/) (2025)
- [An Efficient Adaptive Compression Method for Human Perception and Machine Vision Tasks](https://mlsanity.com/paper/2501.04329/) (2025)
- [M3ISR: A Multi-Modal Multi-View Benchmark for 3D/4D Gaussian Splatting and Feedforward Compression](https://mlsanity.com/paper/2608.22465/) (2026)

## Abstract (arXiv)

Point cloud compression has garnered significant interest in computer vision. However, existing algorithms primarily cater to human vision, while most point cloud data is utilized for machine vision tasks. To address this, we propose a point cloud compression framework that simultaneously handles both human and machine vision tasks. Our framework learns a scalable bit-stream, using only subsets for different machine vision tasks to save bit-rate, while employing the entire bit-stream for human vision tasks. Building on mainstream octree-based frameworks like VoxelContext-Net, OctAttention, and G-PCC, we introduce a new octree depth-level predictor. This predictor adaptively determines the optimal depth level for each octree constructed from a point cloud, controlling the bit-rate for machine vision tasks. For simpler tasks (textite.g., classification) or objects/scenarios, we use fewer depth levels with fewer bits, saving bit-rate. Conversely, for more complex tasks (textite.g., segmentation) or objects/scenarios, we use deeper depth levels with more bits to enhance performance. Experimental results on various datasets (textite.g., ModelNet10, ModelNet40, ShapeNet, ScanNet, and KITTI) show that our point cloud compression approach improves performance for machine vision tasks without compromising human vision quality.