Pixels to Graphs by Associative Embedding

Alejandro Newell, Jia Deng
6/22/2017
cs.CVcs.LG

Official code: princeton-vl/px2graph (136 stars, BSD-3-Clause, last commit 2018-03-26).

Abstract

Graphs are a useful abstraction of image content. Not only can graphs represent details about individual objects in a scene but they can capture the interactions between pairs of objects. We present a method for training a convolutional neural network such that it takes in an input image and produces a full graph definition. This is done end-to-end in a single stage with the use of associative embeddings. The network learns to simultaneously identify all of the elements that make up a graph and piece them together. We benchmark on the Visual Genome dataset, and demonstrate state-of-the-art performance on the challenging task of scene graph generation.

View on arXivView PDF

Code Implementations(1)

princeton-vl/px2graphOfficial100%
BSD-3-Clause

Training code for "Pixels to Graphs by Associative Embedding"

13621PythonDec 22, 20178 years agoBSD-3-Clause

Similar Papers

Image Synthesis with Graph Conditioning: CLIP-Guided Diffusion Models for Scene Graphs

84% match2024

Vision GNN: An Image is Worth Graph of Nodes

84% match1 repo4.4k2022

SelfGraphVQA: A Self-Supervised Graph Neural Network for Scene-based Question Answering

84% match2023

TGraphX: Tensor-Aware Graph Neural Network for Multi-Dimensional Feature Learning

83% match1 repo22025

SGDraw: Scene Graph Drawing Interface Using Object-Oriented Representation

83% match1 repo32022

Weakly-Supervised 3D Scene Graph Generation via Visual-Linguistic Assisted Pseudo-labeling

83% match1 repo92024

Cite this paper

@article{newell2017pixels,
  title  = {Pixels to Graphs by Associative Embedding},
  author = {Alejandro Newell and Jia Deng},
  year   = {2017},
  eprint = {1706.07365},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1706.07365}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1706.07365.svg)](https://codeofpaper.com/paper/1706.07365)

Discussion