Learning the Best Pooling Strategy for Visual Semantic Embedding
Authors:
Preprint
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$\infty$. Without bells and whistles, VSE$\infty$ outperforms previous VSE methods significantly on image-text retrieval benchmarks across popular feature extractors. With a simple adaptation, variants of VSE$\infty$ 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 $\href{https://vse-infty.github.io}{\text{https://vse-infty.github.io}}$.
Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Crisscrossed Captions: Extended Intramodal and Intermodal Semantic Similarity Judgments for MS-COCO2021
- Consensus-Aware Visual-Semantic Embedding for Image-Text Matching2020
- Unicoder-VL: A Universal Encoder for Vision and Language by Cross-Modal Pre-TrainingProceedings of the AAAI Conference on Artificial Intelligence, 2020
- Learning to Represent Image and Text with Denotation Graph2020
- Billion-Scale Similarity Search with GPUsIEEE Transactions on Big Data, 2019
- Fine-Tuning CNN Image Retrieval with No Human AnnotationIEEE Transactions on Pattern Analysis and Machine Intelligence, 2018
- Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image AnnotationsInternational Journal of Computer Vision, 2017
- Spatial Pyramid Pooling in Deep Convolutional Networks for Visual RecognitionIEEE Transactions on Pattern Analysis and Machine Intelligence, 2015
- From image descriptions to visual denotations: New similarity metrics for semantic inference over event descriptionsTransactions of the Association for Computational Linguistics, 2014
- Grounded Compositional Semantics for Finding and Describing Images with SentencesTransactions of the Association for Computational Linguistics, 2014