Multimodal Contrastive Training for Visual Representation Learning

Abstract

We develop an approach to learning visual representations that embraces multimodal data, driven by a combination of intra- and inter-modal similarity preservation objectives. Unlike existing visual pre-training methods, which solve a proxy prediction task in a single domain, our method exploits intrinsic data properties within each modality and semantic information from cross-modal correlation simultaneously, hence improving the quality of learned visual representations. By including multimodal training in a unified framework with different types of contrastive losses, our method can learn more powerful and generic visual features. We first train our model on COCO and evaluate the learned visual representations on various downstream tasks including image classification, object detection, and instance segmentation. For example, the visual representations pre-trained on COCO by our method achieve state-of-the-art top-1 validation accuracy of $55.3\%$ on ImageNet classification, under the common transfer protocol. We also evaluate our method on the large-scale Stock images dataset and show its effectiveness on multi-label image tagging, and cross-modal retrieval tasks.

References in corpus (9)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• A Simple Framework for Contrastive Learning of Visual Representations
• Improved Baselines with Momentum Contrastive Learning
• Learning deep representations by mutual information estimation and maximization
• Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models
• VisualBERT: A Simple and Performant Baseline for Vision and Language
• Data-Efficient Image Recognition with Contrastive Predictive Coding
• Contrastive Multiview Coding
• Self-Supervised MultiModal Versatile Networks

Cited by in corpus (1)

• MIC: Model-agnostic Integrated Cross-channel Recommenders