A Feature-space Multimodal Data Augmentation Technique for Text-video Retrieval

TL;DR

This paper introduces a feature-space multimodal data augmentation method for text-video retrieval that enhances performance by generating new samples through semantic mixing, avoiding raw data transformations and addressing copyright issues.

Contribution

The paper presents a novel feature-space augmentation technique for text-video retrieval that improves accuracy without relying on resource-intensive raw data modifications.

Findings

Significant performance improvements on EPIC-Kitchens-100 dataset

Achieved state-of-the-art results with the proposed method

Conducted extensive ablation studies confirming effectiveness

Abstract

Every hour, huge amounts of visual contents are posted on social media and user-generated content platforms. To find relevant videos by means of a natural language query, text-video retrieval methods have received increased attention over the past few years. Data augmentation techniques were introduced to increase the performance on unseen test examples by creating new training samples with the application of semantics-preserving techniques, such as color space or geometric transformations on images. Yet, these techniques are usually applied on raw data, leading to more resource-demanding solutions and also requiring the shareability of the raw data, which may not always be true, e.g. copyright issues with clips from movies or TV series. To address this shortcoming, we propose a multimodal data augmentation technique which works in the feature space and creates new videos and captions by mixing semantically similar samples. We experiment our solution on a large scale public dataset, EPIC-Kitchens-100, and achieve considerable improvements over a baseline method, improved state-of-the-art performance, while at the same time performing multiple ablation studies. We release code and pretrained models on Github at https://github.com/aranciokov/FSMMDA_VideoRetrieval.