Multi-Agent Reinforcement Learning of 3D Furniture Layout Simulation in Indoor Graphics Scenes

Di, Xinhan; Yu, Po-Lung

doi:10.48550/arxiv.2102.09137

Cited by 3 publications

(3 citation statements)

References 19 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, SceneFormer [33] uses a self-attention mechanism to learn object relations. Di et al [7] use a multi-agent reinforcement learning-based scene design method to learn the optimal 3D furniture layout. With the development of deep generative models [8,20,22,23], indoor scene generation has become another important research direction.…”

Section: Related Work 21 Intelligent Interior Design Systemsmentioning

confidence: 99%

Interactive Interior Design Recommendation via Coarse-to-fine Multimodal Reinforcement Learning

Zhang,

Sun,

Guo

et al. 2023

Proceedings of the 31st ACM International Conference on Multimedia

View full text Add to dashboard Cite

Personalized interior decoration design often incurs high labor costs. Recent efforts in developing intelligent interior design systems have focused on generating textual requirement-based decoration designs while neglecting the problem of how to mine homeowner's hidden preferences and choose the proper initial design. To fill this gap, we propose an Interactive Interior Design Recommendation System (IIDRS) based on reinforcement learning (RL). IIDRS aims to find an ideal plan by interacting with the user, who provides feedback on the gap between the recommended plan and their ideal one. To improve decision-making efficiency and effectiveness in large decoration spaces, we propose a Decoration Recommendation Coarse-to-Fine Policy Network (DecorRCFN). Additionally, to enhance generalization in online scenarios, we propose an objectaware feedback generation method that augments model training with diversified and dynamic textual feedback. Extensive experiments on a real-world dataset demonstrate our method outperforms traditional methods by a large margin in terms of recommendation accuracy. Further user studies demonstrate that our method reaches higher real-world user satisfaction than baseline methods. CCS CONCEPTS• Information systems → Recommender systems.

show abstract

Section: Related Work 21 Intelligent Interior Design Systemsmentioning

confidence: 99%

Interactive Interior Design Recommendation via Coarse-to-fine Multimodal Reinforcement Learning

Zhang,

Sun,

Guo

et al. 2023

Proceedings of the 31st ACM International Conference on Multimedia

View full text Add to dashboard Cite

show abstract

“…Given an initial spatial configuration of boxes, they seek an efficient method to iteratively transport and pack the boxes compactly into a target container. Xinhan et al [32] explore the interior graphics scenes design task as a Markov decision process, which is solved by deep reinforcement learning. Their goal is to generate an accurate layout for the furniture in the indoor graphics scenes simulation.…”

Section: B Reinforcement Learningmentioning

confidence: 99%

Towards Automatic Facility Layout Design Using Reinforcement Learning

Ikeda¹,

Nakagawa²,

Tsuchiya³

2022

Communication Papers of the 17th Conference on Computer Science and Intelligence Systems

View full text Add to dashboard Cite

The accuracy and perfection of layout designing significantly depend on the designer's ability. Quick and nearoptimal designs are very difficult to create. In this study, we propose an automatic design mechanism that can more easily design layouts for various unit groups and sites using reinforcement learning. Specifically, we devised a mechanism to deploy units to be able to fill the largest rectangular space in the current site. We aim to successfully deploy given units within a given site by filling a part of the site. We apply the mechanism to the three sets of units in benchmark problems. The performance was evaluated by changing the learning parameters and iteration count. Consequently, it was possible to produce a layout that successfully deployed units within a given one-floor site.

show abstract

“…Graphic layout generation is emerging as a new research direction for generating realistic and diverse layouts to facilitate design tasks. Recent works show promising methods of layout generation for applications such as graphic user interfaces [2], presentation slides [10], magazines [25], scientific publications [1], commercial advertisements [17,22], Computer-Aided Design (CAD) [24], indoor graphics scenes [4], etc.…”

Section: Introductionmentioning

confidence: 99%

BLT: Bidirectional Layout Transformer for Controllable Layout Generation

Kong¹,

J²,

Chang³

et al. 2021

Preprint

View full text Add to dashboard Cite

Creating visual layouts is an important step in graphic design. Automatic generation of such layouts is important as we seek scale-able and diverse visual designs. Prior works on automatic layout generation focus on unconditional generation, in which the models generate layouts while neglecting user needs for specific problems. To advance conditional layout generation, we introduce BLT, a bidirectional layout transformer. BLT differs from autoregressive decoding as it first generates a draft layout that satisfies the user inputs and then refines the layout iteratively.We verify the proposed model on multiple benchmarks with various fidelity metrics. Our results demonstrate two key advances to the state-of-the-art layout transformer models. First, our model empowers layout transformers to fulfill controllable layout generation. Second, our model slashes the linear inference time in autoregressive decoding into a constant complexity, thereby achieving 4x-10x speedups in generating a layout at inference time.* Work done during their research internship at Google. (a) Conditional (top) and unconditional (bottom) generation.

show abstract

Multi-Agent Reinforcement Learning of 3D Furniture Layout Simulation in Indoor Graphics Scenes

Cited by 3 publications

References 19 publications

Interactive Interior Design Recommendation via Coarse-to-fine Multimodal Reinforcement Learning

Interactive Interior Design Recommendation via Coarse-to-fine Multimodal Reinforcement Learning

Towards Automatic Facility Layout Design Using Reinforcement Learning

BLT: Bidirectional Layout Transformer for Controllable Layout Generation

Contact Info

Product

Resources

About