Multi-Sentence Grounding for Long-term Instructional Video

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Li, Zeqian, Chen, Qirui, Han, Tengda, Zhang, Ya, Wang, Yanfeng, Xie, Weidi
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914878543888384
author Li, Zeqian
Chen, Qirui
Han, Tengda
Zhang, Ya
Wang, Yanfeng
Xie, Weidi
author_facet Li, Zeqian
Chen, Qirui
Han, Tengda
Zhang, Ya
Wang, Yanfeng
Xie, Weidi
contents In this paper, we aim to establish an automatic, scalable pipeline for denoising the large-scale instructional dataset and construct a high-quality video-text dataset with multiple descriptive steps supervision, named HowToStep. We make the following contributions: (i) improving the quality of sentences in dataset by upgrading ASR systems to reduce errors from speech recognition and prompting a large language model to transform noisy ASR transcripts into descriptive steps; (ii) proposing a Transformer-based architecture with all texts as queries, iteratively attending to the visual features, to temporally align the generated steps to corresponding video segments. To measure the quality of our curated datasets, we train models for the task of multi-sentence grounding on it, i.e., given a long-form video, and associated multiple sentences, to determine their corresponding timestamps in the video simultaneously, as a result, the model shows superior performance on a series of multi-sentence grounding tasks, surpassing existing state-of-the-art methods by a significant margin on three public benchmarks, namely, 9.0% on HT-Step, 5.1% on HTM-Align and 1.9% on CrossTask. All codes, models, and the resulting dataset have been publicly released.
format Preprint
id arxiv_https___arxiv_org_abs_2312_14055
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Multi-Sentence Grounding for Long-term Instructional Video
Li, Zeqian
Chen, Qirui
Han, Tengda
Zhang, Ya
Wang, Yanfeng
Xie, Weidi
Computer Vision and Pattern Recognition
In this paper, we aim to establish an automatic, scalable pipeline for denoising the large-scale instructional dataset and construct a high-quality video-text dataset with multiple descriptive steps supervision, named HowToStep. We make the following contributions: (i) improving the quality of sentences in dataset by upgrading ASR systems to reduce errors from speech recognition and prompting a large language model to transform noisy ASR transcripts into descriptive steps; (ii) proposing a Transformer-based architecture with all texts as queries, iteratively attending to the visual features, to temporally align the generated steps to corresponding video segments. To measure the quality of our curated datasets, we train models for the task of multi-sentence grounding on it, i.e., given a long-form video, and associated multiple sentences, to determine their corresponding timestamps in the video simultaneously, as a result, the model shows superior performance on a series of multi-sentence grounding tasks, surpassing existing state-of-the-art methods by a significant margin on three public benchmarks, namely, 9.0% on HT-Step, 5.1% on HTM-Align and 1.9% on CrossTask. All codes, models, and the resulting dataset have been publicly released.
title Multi-Sentence Grounding for Long-term Instructional Video
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.14055