# Planecell: Representing the 3D Space with Planes

## TL;DR

This paper introduces planecell, a novel 3D space representation method that extracts and merges planar segments from stereo images, improving accuracy and efficiency in 3D reconstruction.

## Contribution

The paper presents a new planecell approach that leverages depth-assisted segmentation and CRF-based merging to enhance 3D space representation from stereo images.

## Key findings

- Outperforms existing methods in accuracy on KITTI and Middlebury datasets
- Reduces memory requirements compared to traditional 3D representations
- Enhances reconstruction quality and applicability in 3D modeling

## Abstract

Reconstruction based on the stereo camera has received considerable attention recently, but two particular challenges still remain. The first concerns the need to aggregate similar pixels in an effective approach, and the second is to maintain as much of the available information as possible while ensuring sufficient accuracy. To overcome these issues, we propose a new 3D representation method, namely, planecell, that extracts planarity from the depth-assisted image segmentation and then projects these depth planes into the 3D world. An energy function formulated from Conditional Random Field that generalizes the planar relationships is maximized to merge coplanar segments. We evaluate our method with a variety of reconstruction baselines on both KITTI and Middlebury datasets, and the results indicate the superiorities compared to other 3D space representation methods in accuracy, memory requirements and further applications.