Learning the Best Pooling Strategy for Visual Semantic Embedding

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 https://vse-infty.github.io.

arXiv commentCVPR 2021 camera-ready (oral). The new version fixes a few typos and updates citations

References in corpus (11)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Microsoft COCO Captions: Data Collection and Evaluation Server
• Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
• VisualBERT: A Simple and Performant Baseline for Vision and Language
• Zero-Shot Learning by Convex Combination of Semantic Embeddings
• Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations
• Deep Visual-Semantic Alignments for Generating Image Descriptions
• Unicoder-VL: A Universal Encoder for Vision and Language by Cross-modal Pre-training
• FSPool: Learning Set Representations with Featurewise Sort Pooling
• Crisscrossed Captions: Extended Intramodal and Intermodal Semantic Similarity Judgments for MS-COCO
• Evaluating Text-to-Image Matching using Binary Image Selection (BISON)

Cited by in corpus (4)

• Scaling Up Visual and Vision-Language Representation Learning With Noisy Text Supervision
• Florence: A New Foundation Model for Computer Vision
• Cross-Modal and Uni-Modal Soft-Label Alignment for Image-Text Retrieval
• A Hierarchical Multi-Modal Encoder for Moment Localization in Video Corpus