Neural Motifs: Scene Graph Parsing with Global Context

@article{Zellers2018NeuralMS,
  title={Neural Motifs: Scene Graph Parsing with Global Context},
  author={Rowan Zellers and Mark Yatskar and Sam Thomson and Yejin Choi},
  journal={2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2018},
  pages={5831-5840}
}
  • Rowan Zellers, Mark Yatskar, +1 author Yejin Choi
  • Published 2018
  • Computer Science
  • 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition
  • We investigate the problem of producing structured graph representations of visual scenes. [...] Key Method Our analysis motivates a new baseline: given object detections, predict the most frequent relation between object pairs with the given labels, as seen in the training set. This baseline improves on the previous state-of-the-art by an average of 3.6% relative improvement across evaluation settings.Expand Abstract

    Figures, Tables, and Topics from this paper.

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 161 CITATIONS, ESTIMATED 99% COVERAGE

    Visual Relationships as Functions:Enabling Few-Shot Scene Graph Prediction

    VIEW 8 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Scene Graph Prediction With Limited Labels

    VIEW 7 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Scene Graph Prediction with Limited Labels

    VIEW 7 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Mapping Images to Scene Graphs with Permutation-Invariant Structured Prediction

    VIEW 18 EXCERPTS
    CITES METHODS, RESULTS & BACKGROUND
    HIGHLY INFLUENCED

    Attentive Relational Networks for Mapping Images to Scene Graphs

    VIEW 10 EXCERPTS
    CITES METHODS & BACKGROUND
    HIGHLY INFLUENCED

    LinkNet: Relational Embedding for Scene Graph

    VIEW 6 EXCERPTS
    CITES RESULTS, METHODS & BACKGROUND
    HIGHLY INFLUENCED

    Knowledge-Embedded Routing Network for Scene Graph Generation

    VIEW 9 EXCERPTS
    CITES BACKGROUND & METHODS
    HIGHLY INFLUENCED

    Triplet-Aware Scene Graph Embeddings

    FILTER CITATIONS BY YEAR

    2018
    2020

    CITATION STATISTICS

    • 57 Highly Influenced Citations

    • Averaged 53 Citations per year from 2018 through 2020

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 64 REFERENCES

    Scene Graph Generation from Objects, Phrases and Region Captions

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Scene Graph Generation by Iterative Message Passing

    VIEW 9 EXCERPTS
    HIGHLY INFLUENTIAL

    Detecting Visual Relationships with Deep Relational Networks

    VIEW 2 EXCERPTS

    Graph-Structured Representations for Visual Question Answering

    VIEW 1 EXCERPT

    Deep Variation-Structured Reinforcement Learning for Visual Relationship and Attribute Detection

    VIEW 1 EXCERPT

    Image retrieval using scene graphs

    VIEW 2 EXCERPTS

    Visual Relationship Detection with Internal and External Linguistic Knowledge Distillation

    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL