Segmentation from Natural Language Expressions

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.

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
• Learning Deconvolution Network for Semantic Segmentation
• Ask, Attend and Answer: Exploring Question-Guided Spatial Attention for Visual Question Answering
• Natural Language Object Retrieval

Cited by in corpus (17)

• Multimodal Compact Bilinear Pooling for Visual Question Answering and Visual Grounding
• Learning to Reason: End-to-End Module Networks for Visual Question Answering
• MAttNet: Modular Attention Network for Referring Expression Comprehension
• Person Search with Natural Language Description
• Natural Language Object Retrieval
• VQS: Linking Segmentations to Questions and Answers for Supervised Attention in VQA and Question-Focused Semantic Segmentation
• A Fast and Accurate One-Stage Approach to Visual Grounding
• Weakly-supervised Visual Grounding of Phrases with Linguistic Structures
• Utilizing Large Scale Vision and Text Datasets for Image Segmentation from Referring Expressions
• Multi-modal Capsule Routing for Actor and Action Video Segmentation Conditioned on Natural Language Queries
• Weakly Supervised Person Re-ID: Differentiable Graphical Learning and A New Benchmark
• Learning to discover and localize visual objects with open vocabulary
• Discover and Learn New Objects from Documentaries
• Recurrent Instance Segmentation using Sequences of Referring Expressions
• Recurrent Multimodal Interaction for Referring Image Segmentation
• Learning Action Concept Trees and Semantic Alignment Networks from Image-Description Data
• Resolving Referring Expressions in Images With Labeled Elements