Pixels to Graphs by Associative Embedding

06/22/2017
by   Alejandro Newell, et al.
0

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. 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 report a Recall@50 of 9.7 state-of-the-art at 3.4 task of scene graph generation.

READ FULL TEXT

page 2

page 7

research
01/10/2017

Scene Graph Generation by Iterative Message Passing

Understanding a visual scene goes beyond recognizing individual objects ...
research
09/23/2021

Scene Graph Generation for Better Image Captioning?

We investigate the incorporation of visual relationships into the task o...
research
05/13/2020

Structured Query-Based Image Retrieval Using Scene Graphs

A structured query can capture the complexity of object interactions (e....
research
04/08/2019

End-to-end Projector Photometric Compensation

Projector photometric compensation aims to modify a projector input imag...
research
02/15/2018

Mapping Images to Scene Graphs with Permutation-Invariant Structured Prediction

Structured prediction is concerned with predicting multiple inter-depend...
research
06/04/2021

A Deep Local and Global Scene-Graph Matching for Image-Text Retrieval

Conventional approaches to image-text retrieval mainly focus on indexing...
research
12/06/2018

Scene Dynamics: Counterfactual Critic Multi-Agent Training for Scene Graph Generation

Scene graphs -- objects as nodes and visual relationships as edges -- de...

Please sign up or login with your details

Forgot password? Click here to reset