MonteFloor: Extending MCTS for Reconstructing Accurate Large-Scale Floor Plans
This work addresses the challenge of generating precise vectorized floor plans for large complex scenes, which is important for applications in architecture and robotics, though it builds incrementally on existing projection and proposal extraction techniques.
The authors tackled the problem of reconstructing accurate floor plans from noisy 3D point clouds by extending Monte Carlo Tree Search (MCTS) to optimize room proposals, achieving significant improvements over state-of-the-art methods on Structured3D and Floor-SP datasets.
We propose a novel method for reconstructing floor plans from noisy 3D point clouds. Our main contribution is a principled approach that relies on the Monte Carlo Tree Search (MCTS) algorithm to maximize a suitable objective function efficiently despite the complexity of the problem. Like previous work, we first project the input point cloud to a top view to create a density map and extract room proposals from it. Our method selects and optimizes the polygonal shapes of these room proposals jointly to fit the density map and outputs an accurate vectorized floor map even for large complex scenes. To do this, we adapted MCTS, an algorithm originally designed to learn to play games, to select the room proposals by maximizing an objective function combining the fitness with the density map as predicted by a deep network and regularizing terms on the room shapes. We also introduce a refinement step to MCTS that adjusts the shape of the room proposals. For this step, we propose a novel differentiable method for rendering the polygonal shapes of these proposals. We evaluate our method on the recent and challenging Structured3D and Floor-SP datasets and show a significant improvement over the state-of-the-art, without imposing any hard constraints nor assumptions on the floor plan configurations.