DIBS: Enhancing Dense Video Captioning with Unlabeled Videos via Pseudo Boundary Enrichment and Online Refinement

TL;DR

DIBS introduces a pretraining framework for dense video captioning that utilizes unlabeled videos and large language models to improve caption quality and boundary detection through pseudo labels and online refinement, achieving state-of-the-art results.

Contribution

The paper proposes a novel pretraining method that leverages unlabeled videos and large language models, with an online boundary refinement strategy, to enhance dense video captioning performance.

Findings

Outperforms previous state-of-the-art on YouCook2 and ActivityNet datasets.

Achieves significant improvements using only 0.4% of unlabeled data compared to prior methods.

Effectively refines pseudo boundaries during training for better captioning accuracy.

Abstract

We present Dive Into the BoundarieS (DIBS), a novel pretraining framework for dense video captioning (DVC), that elaborates on improving the quality of the generated event captions and their associated pseudo event boundaries from unlabeled videos. By leveraging the capabilities of diverse large language models (LLMs), we generate rich DVC-oriented caption candidates and optimize the corresponding pseudo boundaries under several meticulously designed objectives, considering diversity, event-centricity, temporal ordering, and coherence. Moreover, we further introduce a novel online boundary refinement strategy that iteratively improves the quality of pseudo boundaries during training. Comprehensive experiments have been conducted to examine the effectiveness of the proposed technique components. By leveraging a substantial amount of unlabeled video data, such as HowTo100M, we achieve a remarkable advancement on standard DVC datasets like YouCook2 and ActivityNet. We outperform the previous state-of-the-art Vid2Seq across a majority of metrics, achieving this with just 0.4% of the unlabeled video data used for pre-training by Vid2Seq.