Papers › Multimodal Pretraining for Dense Video Captioning

Multimodal Pretraining for Dense Video Captioning

10 Nov 2020Asian Chapter of the Association for Computational Linguistics 2020arXiv:2011.11760archive 2025-07-28

Gabriel Huang, Bo Pang, Zhenhai Zhu, Clara Rivera, Radu Soricut

Learning specific hands-on skills such as cooking, car maintenance, and home repairs increasingly happens via instructional videos. The user experience with such videos is known to be improved by meta-information such as time-stamped annotations for the main steps involved. Generating such annotations automatically is challenging, and we describe here two relevant contributions. First, we construct and release a new dense video captioning dataset, Video Timeline Tags (ViTT), featuring a variety of instructional videos together with time-stamped annotations. Second, we explore several multimodal sequence-to-sequence pretraining strategies that leverage large unsupervised datasets of videos and caption-like texts. We pretrain and subsequently finetune dense video captioning models using both YouCook2 and ViTT. We show that such models generalize well and are robust over a wide variety of instructional videos.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

google-research-datasets/Video-Timeline-Tags-ViTT officialmentioned in papermentioned on GitHub report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Dense Video CaptioningVideo Captioning

Datasets

Introduced by this paper, per the archive.

ViTT

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Dense Video Captioning YouCook2 E2vidD6-MASSalign-BiD ROUGE-L 39.03 #7 of 7 Archive leaderboard report
Video Captioning YouCook2 E2vidD6-MASSvid-BiD BLEU-4 12.04 #6 of 14 Archive leaderboard report
Video Captioning YouCook2 E2vidD6-MASSvid-BiD CIDEr 1.22 #6 of 14 Archive leaderboard report
Video Captioning YouCook2 E2vidD6-MASSvid-BiD METEOR 18.32 #6 of 14 Archive leaderboard report
Video Captioning YouCook2 E2vidD6-MASSvid-BiD ROUGE-L 39.03 #6 of 14 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections