mlsanityData to 7 Oct 2026

Unsupervised Traffic Scene Generation with Synthetic 3D Scene Graphs

arXiv 2303.08473 ↗First posted 15 Mar 2023cs.CV cs.LG cs.ROcited by 2

Not on the map. Why?

Abstract (arXiv)

Image synthesis driven by computer graphics achieved recently a remarkable realism, yet synthetic image data generated this way reveals a significant domain gap with respect to real-world data. This is especially true in autonomous driving scenarios, which represent a critical aspect for overcoming utilizing synthetic data for training neural networks. We propose a method based on domain-invariant scene representation to directly synthesize traffic scene imagery without rendering. Specifically, we rely on synthetic scene graphs as our internal representation and introduce an unsupervised neural network architecture for realistic traffic scene synthesis. We enhance synthetic scene graphs with spatial information about the scene and demonstrate the effectiveness of our approach through scene manipulation.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 29 of its 39 references that are in the map. First posted on arXiv 15 Mar 2023.

Builds on

29 of 39 references.

References · 29

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2016Image-to-Image Translation with Conditional Adversarial Networkscited by 4,510
  3. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  4. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,580
  5. 2016Improved Techniques for Training GANscited by 2,938
  6. 2016Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotationscited by 2,646
  7. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  8. 2017CARLA: An Open Urban Driving Simulatorcited by 2,431
  9. 2017Improved Training of Wasserstein GANscited by 2,223
  10. 2019Semantic Image Synthesis with Spatially-Adaptive Normalizationcited by 1,295
  11. 2018BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learningcited by 1,282
  12. 2017High-Resolution Image Synthesis and Semantic Manipulation with Conditional GANscited by 1,176
17 more
  1. 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,128
  2. 2018Multimodal Unsupervised Image-to-Image Translationcited by 795
  3. 2020Contrastive Learning for Unpaired Image-to-Image Translationcited by 637
  4. 2016StackGAN: Text to Photo-realistic Image Synthesis with Stacked Generative Adversarial Networkscited by 575
  5. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
  6. 2017AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehiclescited by 517
  7. 2016Virtual Worlds as Proxy for Multi-Object Tracking Analysiscited by 340
  8. 2018Image Generation from Scene Graphscited by 324
  9. 2017DualGAN: Unsupervised Dual Learning for Image-to-Image Translationcited by 310
  10. 2017Photographic Image Synthesis with Cascaded Refinement Networkscited by 223
  11. 2018A DIRT-T Approach to Unsupervised Domain Adaptationcited by 215
  12. 2017Playing for Benchmarkscited by 169
  13. 2018Synscapes: A Photorealistic Synthetic Dataset for Street Scene Parsingcited by 105
  14. 2020Semantic Image Manipulation Using Scene Graphscited by 82
  15. 2018Augmented CycleGAN: Learning Many-to-Many Mappings from Unpaired Datacited by 76
  16. 2019Specifying Object Attributes and Relations in Interactive Scene Generationcited by 76
  17. 2017Pixels to Graphs by Associative Embeddingcited by 58

Built on by

Cited by year

Citing papers per year: 2024: 2.

2 papers.

  1. 2024SceneCraft: An LLM Agent for Synthesizing 3D Scene as Blender Codecited by 125
  2. 2024Automatic Scene Generation: State-of-the-Art Techniques, Models, Datasets, Challenges, and Future Prospectscited by 2