Revisiting Image-Language Networks for Open-ended Phrase Detection
arXiv:1811.07212 · doi:10.1109/TPAMI.2020.3029008
Abstract
Most existing work that grounds natural language phrases in images starts with the assumption that the phrase in question is relevant to the image. In this paper we address a more realistic version of the natural language grounding task where we must both identify whether the phrase is relevant to an image and localize the phrase. This can also be viewed as a generalization of object detection to an open-ended vocabulary, introducing elements of few- and zero-shot detection. We propose an approach for this task that extends Faster R-CNN to relate image regions and phrases. By carefully initializing the classification layers of our network using canonical correlation analysis (CCA), we encourage a solution that is more discerning when reasoning between similar phrases, resulting in over double the performance compared to a naive adaptation on three popular phrase grounding datasets, Flickr30K Entities, ReferIt Game, and Visual Genome, with test-time phrase vocabulary sizes of 5K, 32K, and 159K, respectively.
arXiv commentAccepted to TPAMI
References in corpus (12)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Efficient Estimation of Word Representations in Vector Space
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
- Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
- Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations
- Grounding of Textual Phrases in Images by Reconstruction
- Generation and Comprehension of Unambiguous Object Descriptions
- Deep Image Retrieval: Learning global representations for image search
- Attention Correctness in Neural Image Captioning
- Learning Two-Branch Neural Networks for Image-Text Matching Tasks
- VQD: Visual Query Detection in Natural Scenes