Segmentation from Natural Language Expressions

TL;DR

This paper introduces a novel end-to-end neural network model that segments images based on natural language expressions, enabling more flexible and precise image segmentation compared to traditional class-based methods.

Contribution

The paper presents a new neural network architecture combining recurrent and convolutional components for language-guided image segmentation, outperforming previous fixed-category approaches.

Findings

The model achieves high-quality segmentation from natural language expressions.

It significantly outperforms baseline methods on benchmark datasets.

The approach effectively combines visual and linguistic information in an end-to-end framework.

Abstract

In this paper we approach the novel problem of segmenting an image based on a natural language expression. This is different from traditional semantic segmentation over a predefined set of semantic classes, as e.g., the phrase "two men sitting on the right bench" requires segmenting only the two people on the right bench and no one standing or sitting on another bench. Previous approaches suitable for this task were limited to a fixed set of categories and/or rectangular regions. To produce pixelwise segmentation for the language expression, we propose an end-to-end trainable recurrent and convolutional network model that jointly learns to process visual and linguistic information. In our model, a recurrent LSTM network is used to encode the referential expression into a vector representation, and a fully convolutional network is used to a extract a spatial feature map from the image and output a spatial response map for the target object. We demonstrate on a benchmark dataset that our model can produce quality segmentation output from the natural language expression, and outperforms baseline methods by a large margin.