Monte Carlo Scene Search for 3D Scene Understanding

TL;DR

This paper introduces a modified Monte Carlo Tree Search algorithm for 3D scene understanding from noisy RGB-D data, enabling efficient analysis-by-synthesis without extensive training data, and demonstrates superior layout retrieval on ScanNet.

Contribution

It adapts MCTS for perception tasks, reducing reliance on training data and hyperparameter tuning, and improves 3D scene analysis from RGB-D scans.

Findings

Often retrieves configurations better than manual annotations.

Effective in optimizing object and layout hypotheses.

Demonstrates applicability on the ScanNet dataset.

Abstract

We explore how a general AI algorithm can be used for 3D scene understanding to reduce the need for training data. More exactly, we propose a modification of the Monte Carlo Tree Search (MCTS) algorithm to retrieve objects and room layouts from noisy RGB-D scans. While MCTS was developed as a game-playing algorithm, we show it can also be used for complex perception problems. Our adapted MCTS algorithm has few easy-to-tune hyperparameters and can optimise general losses. We use it to optimise the posterior probability of objects and room layout hypotheses given the RGB-D data. This results in an analysis-by-synthesis approach that explores the solution space by rendering the current solution and comparing it to the RGB-D observations. To perform this exploration even more efficiently, we propose simple changes to the standard MCTS' tree construction and exploration policy. We demonstrate our approach on the ScanNet dataset. Our method often retrieves configurations that are better than some manual annotations, especially on layouts.