Corpus ID: 54200214

SPLAT: Semantic Pixel-Level Adaptation Transforms for Detection

@article{Tzeng2018SPLATSP,
  title={SPLAT: Semantic Pixel-Level Adaptation Transforms for Detection},
  author={E. Tzeng and Kaylee Burns and Kate Saenko and Trevor Darrell},
  journal={ArXiv},
  year={2018},
  volume={abs/1812.00929}
}
  • E. Tzeng, Kaylee Burns, +1 author Trevor Darrell
  • Published 2018
  • Computer Science
  • ArXiv
  • Domain adaptation of visual detectors is a critical challenge, yet existing methods have overlooked pixel appearance transformations, focusing instead on bootstrapping and/or domain confusion losses. We propose a Semantic Pixel-Level Adaptation Transform (SPLAT) approach to detector adaptation that efficiently generates cross-domain image pairs. Our model uses aligned-pair and/or pseudo-label losses to adapt an object detector to the target domain, and can learn transformations with or without… CONTINUE READING
    10 Citations
    SCL: TOWARDS ACCURATE DOMAIN ADAPTIVE OB-
    • 2019
    • PDF
    Domain Adaptation for Object Detection via Style Consistency
    • 13
    • PDF
    NoVA: Learning to See in Novel Viewpoints and Domains
    • 1
    • Highly Influenced
    • PDF
    Unbiased Mean Teacher for Cross Domain Object Detection
    • PDF
    A Review of Single-Source Deep Unsupervised Visual Domain Adaptation
    • 3
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 36 REFERENCES
    Domain Adaptive Faster R-CNN for Object Detection in the Wild
    • 268
    • Highly Influential
    • PDF
    Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
    • 808
    • PDF
    CyCADA: Cycle-Consistent Adversarial Domain Adaptation
    • 1,020
    • PDF
    Semi-supervised Domain Adaptation with Instance Constraints
    • 126
    • PDF
    Adapting Visual Category Models to New Domains
    • 1,486
    • PDF
    Fully Convolutional Networks for Semantic Segmentation
    • 8,532
    • PDF
    Feature Pyramid Networks for Object Detection
    • 4,976
    • PDF
    The Cityscapes Dataset for Semantic Urban Scene Understanding
    • 3,705
    • Highly Influential
    • PDF
    Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
    • 20,974
    • PDF
    Unsupervised Domain Adaptation by Backpropagation
    • 1,903
    • PDF