Corpus ID: 219636196

SegNBDT: Visual Decision Rules for Segmentation

  title={SegNBDT: Visual Decision Rules for Segmentation},
  author={Alvin Wan and D. Ho and Younjin Song and Henk Tillman and S. A. Bargal and J. Gonzalez},
  • Alvin Wan, D. Ho, +3 authors J. Gonzalez
  • Published 2020
  • Computer Science
  • ArXiv
  • The black-box nature of neural networks limits model decision interpretability, in particular for high-dimensional inputs in computer vision and for dense pixel prediction tasks like segmentation. To address this, prior work combines neural networks with decision trees. However, such models (1) perform poorly when compared to state-of-the-art segmentation models or (2) fail to produce decision rules with spatially-grounded semantic meaning. In this work, we build a hybrid neural-network and… CONTINUE READING


    Neural Decision Forests for Semantic Image Labelling
    • 102
    • PDF
    Convolutional Decision Trees for Feature Learning and Segmentation
    • 14
    • PDF
    Deep Neural Decision Forests
    • 280
    • PDF
    Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
    • 2,234
    • Highly Influential
    • PDF
    RISE: Randomized Input Sampling for Explanation of Black-box Models
    • 161
    • Highly Influential
    • PDF
    NBDT: Neural-Backed Decision Trees
    • 11
    • PDF
    Fast Residual Forests: Rapid Ensemble Learning for Semantic Segmentation
    • 13
    • PDF
    Interpreting CNNs via Decision Trees
    • Q. Zhang, Yu Yang, Y. Wu, S. Zhu
    • Computer Science
    • 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
    • 2019
    • 114
    • PDF