CVCLNov 3, 2022

Zero-shot Video Moment Retrieval With Off-the-Shelf Models

arXiv:2211.02178v17 citationsh-index: 88
Originality Incremental advance
AI Analysis

This addresses the bottleneck of expensive data annotation and limited compute for building models on new tasks, offering a practical solution for the machine learning community, though it is incremental as it builds on existing models.

The paper tackles the problem of Video Moment Retrieval (VMR) by proposing a zero-shot approach that repurposes off-the-shelf models without finetuning, achieving at least a 2.5x improvement over previous zero-shot methods and reducing the gap to state-of-the-art supervised models by over 74% on the QVHighlights benchmark.

For the majority of the machine learning community, the expensive nature of collecting high-quality human-annotated data and the inability to efficiently finetune very large state-of-the-art pretrained models on limited compute are major bottlenecks for building models for new tasks. We propose a zero-shot simple approach for one such task, Video Moment Retrieval (VMR), that does not perform any additional finetuning and simply repurposes off-the-shelf models trained on other tasks. Our three-step approach consists of moment proposal, moment-query matching and postprocessing, all using only off-the-shelf models. On the QVHighlights benchmark for VMR, we vastly improve performance of previous zero-shot approaches by at least 2.5x on all metrics and reduce the gap between zero-shot and state-of-the-art supervised by over 74%. Further, we also show that our zero-shot approach beats non-pretrained supervised models on the Recall metrics and comes very close on mAP metrics; and that it also performs better than the best pretrained supervised model on shorter moments. Finally, we ablate and analyze our results and propose interesting future directions.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes