Learning the Best Pooling Strategy for Visual Semantic Embedding
Authors:
Top Cited Papers
- 1 June 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 15784-15793
Abstract
Visual Semantic Embedding (VSE) is a dominant approach for vision-language retrieval, which aims at learning a deep embedding space such that visual data are embedded close to their semantic text labels or descriptions. Recent VSE models use complex methods to better contextualize and aggregate multi-modal features into holistic embeddings. However, we discover that surprisingly simple (but carefully selected) global pooling functions (e.g., max pooling) outperform those complex models, across different feature extractors. Despite its simplicity and effectiveness, seeking the best pooling function for different data modality and feature extractor is costly and tedious, especially when the size of features varies (e.g., text, video). Therefore, we propose a Generalized Pooling Operator (GPO), which learns to automatically adapt itself to the best pooling strategy for different features, requiring no manual tuning while staying effective and efficient. We extend the VSE model using this proposed GPO and denote it as VSE∞.Without bells and whistles, VSE∞ outperforms previous VSE methods significantly on image-text retrieval benchmarks across popular feature extractors. With a simple adaptation, variants of VSE∞ further demonstrate its strength by achieving the new state of the art on two video-text retrieval datasets. Comprehensive experiments and visualizations confirm that GPO always discovers the best pooling strategy and can be a plug-and-play feature aggregation module for standard VSE models. Code and pre-trained models are available at http://jcchen.me/vse_infty/Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Dual Attention Networks for Multimodal Reasoning and Matching2017
- Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image AnnotationsInternational Journal of Computer Vision, 2017
- Synthesized Classifiers for Zero-Shot Learning2016
- MSR-VTT: A Large Video Description Dataset for Bridging Video and Language2016
- Deep visual-semantic alignments for generating image descriptions2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- A Convolutional Neural Network for Modelling Sentences2014
- Attribute-Based Classification for Zero-Shot Visual Object CategorizationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Long Short-Term MemoryNeural Computation, 1997