MAD: A Scalable Dataset for Language Grounding in Videos from Movie Audio Descriptions
This provides a more challenging benchmark for video-language grounding research, addressing dataset limitations that cause overfitting in state-of-the-art techniques.
The authors tackled the problem of hidden biases in video-language grounding datasets by introducing MAD, a scalable dataset with over 384,000 sentences grounded in 1,200 hours of videos, which reduces these biases and enables grounding in long-form videos up to three hours.
The recent and increasing interest in video-language research has driven the development of large-scale datasets that enable data-intensive machine learning techniques. In comparison, limited effort has been made at assessing the fitness of these datasets for the video-language grounding task. Recent works have begun to discover significant limitations in these datasets, suggesting that state-of-the-art techniques commonly overfit to hidden dataset biases. In this work, we present MAD (Movie Audio Descriptions), a novel benchmark that departs from the paradigm of augmenting existing video datasets with text annotations and focuses on crawling and aligning available audio descriptions of mainstream movies. MAD contains over 384,000 natural language sentences grounded in over 1,200 hours of videos and exhibits a significant reduction in the currently diagnosed biases for video-language grounding datasets. MAD's collection strategy enables a novel and more challenging version of video-language grounding, where short temporal moments (typically seconds long) must be accurately grounded in diverse long-form videos that can last up to three hours. We have released MAD's data and baselines code at https://github.com/Soldelli/MAD.