Monte Carlo Scene Search for 3D Scene Understanding

Shreyas Hampali Shivakumar, Sinisa Stekovic, Sayan Deb Sarkar, Chetan Srinivasa Kumar, Friedrich Fraundorfer, Vincent Lepetit

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandBegutachtung

Abstract

We explore how a general AI algorithm can be used for 3D scene understanding in order 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. It 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.
Originalspracheenglisch
Titel2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Seiten13799-13808
ISBN (elektronisch)978-1-6654-4509-2
DOIs
PublikationsstatusVeröffentlicht - 2021
Veranstaltung2021 IEEE Conference on Computer Vision and Pattern Recognition: CVPR 2021 - Virtuell
Dauer: 19 Juni 202125 Juni 2021

Konferenz

Konferenz2021 IEEE Conference on Computer Vision and Pattern Recognition
KurztitelCVPR 2021
OrtVirtuell
Zeitraum19/06/2125/06/21

Fingerprint

Untersuchen Sie die Forschungsthemen von „Monte Carlo Scene Search for 3D Scene Understanding“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren