paper-with-me

홈 › Papers

Unconditional Scene Graph Generation

2021-08-12 · ICCV 2021 10 · Sarthak Garg, Helisa Dhamo, Azade Farshad, Sabrina Musatian, Nassir Navab, Federico Tombari

Despite recent advancements in single-domain or single-object image generation, it is still challenging to generate complex scenes containing diverse, multiple objects and their interactions. Scene graphs, composed of nodes as objects and directed-edges as relationships among objects, offer an alternative representation of a scene that is more semantically grounded than images. We hypothesize that a generative model for scene graphs might be able to learn the underlying semantic structure of real-world scenes more effectively than images, and hence, generate realistic novel scenes in the form of scene graphs. In this work, we explore a new task for the unconditional generation of semantic scene graphs. We develop a deep auto-regressive model called SceneGraphGen which can directly learn the probability distribution over labelled and directed graphs using a hierarchical recurrent architecture. The model takes a seed object as input and generates a scene graph in a sequence of steps, each step generating an object node, followed by a sequence of relationship edges connecting to the previous nodes. We show that the scene graphs generated by SceneGraphGen are diverse and follow the semantic patterns of real-world scenes. Additionally, we demonstrate the application of the generated graphs in image synthesis, anomaly detection and scene graph completion.

📄 PDF Abstract BibTeX arXiv:2108.05884

Code (0)

등록된 구현이 없습니다.

Tasks

Anomaly DetectionGraph GenerationImage GenerationObjectScene Graph Generation

Similar Papers 제목 키워드 기반

End-to-End Optimization of Scene Layout

2020-07-23 · CVPR 2020 6 · Andrew Luo, Zhoutong Zhang, Jiajun Wu, Joshua B. Tenenbaum

We propose an end-to-end variational generative model for scene layout synthesis conditioned on scene graphs. Unlike unconditional scene layout generation, we use scene graphs as an abstract but general representation to…

DiversityIndoor Scene ReconstructionIndoor Scene SynthesisLayout Generation+2

Unconditional Synthesis of Complex Scenes Using a Semantic Bottleneck

2021-01-01 · Samaneh Azadi, Michael Tschannen, Eric Tzeng, Sylvain Gelly 외

Coupling the high-fidelity generation capabilities of label-conditional image synthesis methods with the flexibility of unconditional generative models, we propose a semantic bottleneck GAN model for unconditional synthe…

Image GenerationSegmentation

Joint Generative Modeling of Scene Graphs and Images via Diffusion Models

2024-01-02 · Bicheng Xu, Qi Yan, Renjie Liao, Lele Wang 외

In this paper, we present a novel generative task: joint scene graph - image generation. While previous works have explored image generation conditioned on scene graphs or layouts, our task is distinctive and important a…

Graph GenerationImage GenerationScene Graph DetectionScene Graph Generation

Towards Improving the Generation Quality of Autoregressive Slot VAEs

2022-06-03 · Patrick Emami, Pan He, Sanjay Ranka, Anand Rangarajan

Unconditional scene inference and generation are challenging to learn jointly with a single compositional model. Despite encouraging progress on models that extract object-centric representations (''slots'') from images,…

Image GenerationObjectScene GenerationScene Understanding+1

Semantic Bottleneck Scene Generation

2019-11-26 · Samaneh Azadi, Michael Tschannen, Eric Tzeng, Sylvain Gelly 외

Coupling the high-fidelity generation capabilities of label-conditional image synthesis methods with the flexibility of unconditional generative models, we propose a semantic bottleneck GAN model for unconditional synthe…

Conditional Image GenerationImage GenerationImage-to-Image TranslationScene Generation+1