mlsanityData to 7 Oct 2026

KLIEP-based Density Ratio Estimation for Semantically Consistent Synthetic to Real Images Adaptation in Urban Traffic Scenes

arXiv 2105.12549 ↗First posted 26 May 2021cs.CVcited by 1

Not on the map. Why?

Abstract (arXiv)

Synthetic data has been applied in many deep learning based computer vision tasks. Limited performance of algorithms trained solely on synthetic data has been approached with domain adaptation techniques such as the ones based on generative adversarial framework. We demonstrate how adversarial training alone can introduce semantic inconsistencies in translated images. To tackle this issue we propose density prematching strategy using KLIEP-based density ratio estimation procedure. Finally, we show that aforementioned strategy improves quality of translated images of underlying method and their usability for the semantic segmentation task in the context of autonomous driving.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 31 of its 53 references that are in the map. First posted on arXiv 26 May 2021.

Builds on

31 of 53 references.

References · 31

  1. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  2. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  3. 2017CyCADA: Cycle-Consistent Adversarial Domain Adaptationcited by 1,079
  4. 2015FlowNet: Learning Optical Flow with Convolutional Networkscited by 1,054
  5. 2016Playing for Data: Ground Truth from Computer Gamescited by 816
  6. 2017Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networkscited by 707
  7. 2017Maximum Classifier Discrepancy for Unsupervised Domain Adaptationcited by 682
  8. 2017Unsupervised Image-to-Image Translation Networkscited by 602
  9. 2017Self-Normalizing Neural Networkscited by 591
  10. 2018Learning to Adapt Structured Output Space for Semantic Segmentationcited by 584
  11. 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 552
  12. 2017Learning from Synthetic Humanscited by 358
19 more
  1. 2016Virtual Worlds as Proxy for Multi-Object Tracking Analysiscited by 340
  2. 2017DualGAN: Unsupervised Dual Learning for Image-to-Image Translationcited by 310
  3. 2018Taking A Closer Look at Domain Shift: Category-level Adversaries for Semantics Consistent Domain Adaptationcited by 295
  4. 2017Dilated Residual Networkscited by 293
  5. 2016Learning from Simulated and Unsupervised Images through Adversarial Trainingcited by 287
  6. 2016FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptationcited by 242
  7. 2016Coupled Generative Adversarial Networkscited by 216
  8. 2018A DIRT-T Approach to Unsupervised Domain Adaptationcited by 215
  9. 2016Unsupervised Cross-Domain Image Generationcited by 125
  10. 2018Geometry-Consistent Generative Adversarial Networks for One-Sided Unsupervised Domain Mappingcited by 113
  11. 2020CrDoCo: Pixel-level Domain Transfer with Cross-Domain Consistencycited by 107
  12. 2017ROAD: Reality Oriented Adaptation for Semantic Segmentation of Urban Scenescited by 95
  13. 2019Sliced Wasserstein Generative Modelscited by 61
  14. 2018Effective Use of Synthetic Data for Urban Scene Semantic Segmentationcited by 41
  15. 2015SceneNet: Understanding Real World Indoor Scenes With Synthetic Datacited by 29
  16. 2017Boundary-Seeking Generative Adversarial Networkscited by 27
  17. 2018Semantic-aware Grad-GAN for Virtual-to-Real Urban Scene Adaptioncited by 27
  18. 2015Semantic Pose using Deep Networks Trained on Synthetic RGB-Dcited by 6
  19. 2017Non-parametric estimation of Jensen-Shannon Divergence in Generative Adversarial Network trainingcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Flow-Based Density Ratio Estimation for Intractable Distributions with Applications in Genomicscited by 0