Monte Carlo Scene Search for 3D Scene Understanding
arXiv:2103.07969
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.
arXiv commentTo be presented at CVPR 2021
References in corpus (6)
- Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Images
- LayoutNet: Reconstructing the 3D Room Layout from a Single RGB Image
- Structured3D: A Large Photo-realistic Dataset for Structured 3D Modeling
- HorizonNet: Learning Room Layout with 1D Representation and Pano Stretch Data Augmentation
- Mask2CAD: 3D Shape Prediction by Learning to Segment and Retrieve
- GeoLayout: Geometry Driven Room Layout Estimation Based on Depth Maps of Planes