Efficient Plane Segmentation in Depth Image Based on Adaptive Patch-Wise Region Growing
Lantao Zhang, Haochen Niu, Peilin Liu, Fei Wen, RENDONG YING
AI summary
Problem
Existing plane segmentation methods struggle with computational redundancy on varying planar region sizes or require excessive resources, making real-time deployment on resource-limited edge devices challenging.
Approach
APRG constructs a data pyramid from depth images to adaptively extract variable-sized planar patches in a top-down manner, then merges them using a specialized patch-wise region growing algorithm for efficient boundary refinement.
Key results
- Achieves over 600 FPS at 640x480 resolution on a mid-range CPU
- Outperforms state-of-the-art methods by a factor of 1.46 in speed
- Significantly improves segmentation accuracy on noisy real-world depth data
- Introduces a novel adaptive-patch representation that eliminates redundant plane fitting
Why it matters
Enables real-time, high-precision 3D scene understanding for resource-constrained robotic systems and edge devices.
Abstract
Plane segmentation algorithms are widely used in robotics, serving key roles in scenarios such as indoor local- ization, scene understanding, and robotic manipulation. These applications typically require real-time, precise, and robust plane segmentation processing, which presents a significant challenge. Existing methods based on pixel-wise or fix-sized patch-wise operation are redundant, as planar regions in real-world scenes are of diverse sizes. In this paper, we introduce a highly efficient method for plane segmentation, namely Adaptive Patch-wise Region Growing (APRG). APRG begins with data sampling to construct a data pyramid. To avoid redundant planer fitting in large planar regions, we introduce an adaptive patch-wise plane fitting algorithm with the pyramid accessed in a top-down manner. The largest possible planar patches are obtained in this process. Subsequently we introduce a region growing algorithm specially designed for our patch representation. Overall, APRG achieves more than 600 FPS at a 640x480 resolution on a mid-range CPU without using parallel acceleration techniques, which outperforms the state-of-the-art method by a factor of 1.46. Besides, in addition to its speedup in run-time, APRG significantly improves the segmentation quality, especially on real-world data. Code is available at ZhangLanTao/APRG.