Papers › Cap4Video: What Can Auxiliary Captions Do for Text-Video Retrieval?

Cap4Video: What Can Auxiliary Captions Do for Text-Video Retrieval?

31 Dec 2022CVPR 2023 1arXiv:2301.00184archive 2025-07-28

Wenhao Wu, Haipeng Luo, Bo Fang, Jingdong Wang, Wanli Ouyang

Most existing text-video retrieval methods focus on cross-modal matching between the visual content of videos and textual query sentences. However, in real-world scenarios, online videos are often accompanied by relevant text information such as titles, tags, and even subtitles, which can be utilized to match textual queries. This insight has motivated us to propose a novel approach to text-video retrieval, where we directly generate associated captions from videos using zero-shot video captioning with knowledge from web-scale pre-trained models (e.g., CLIP and GPT-2). Given the generated captions, a natural question arises: what benefits do they bring to text-video retrieval? To answer this, we introduce Cap4Video, a new framework that leverages captions in three ways: i) Input data: video-caption pairs can augment the training data. ii) Intermediate feature interaction: we perform cross-modal feature interaction between the video and caption to produce enhanced video representations. iii) Output score: the Query-Caption matching branch can complement the original Query-Video matching branch for text-video retrieval. We conduct comprehensive ablation studies to demonstrate the effectiveness of our approach. Without any post-processing, Cap4Video achieves state-of-the-art performance on four standard text-video retrieval benchmarks: MSR-VTT (51.4%), VATEX (66.6%), MSVD (51.8%), and DiDeMo (52.0%). The code is available at https://github.com/whwu95/Cap4Video .

PaperPDFConference PDFCodeCode Syntology ran

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

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2301.00184")

Code

Syntology Ran 14 of 25 code samples harvested from 2 repositories linked to this paper; 11 have no recorded run. Of those that ran: 2 ran · honoured contract; 1 ran · violated contract; 7 ran · our draft was wrong; 1 ran · fixture could not drive it; 3 ran with no contract checked.

By repository: official repository: 15 samples from 1 repository, 8 ran; community (archive-listed): 10 samples from 1 repository, 6 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

whwu95/Cap4Video officialmentioned in papermentioned on GitHubpytorchMIT report
whwu95/BIKE mentioned on GitHubpytorchMIT report
whwu95/GPT4Vis mentioned on GitHubMIT report
whwu95/text4vis mentioned on GitHubpytorchMIT 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

25 samples harvested; 14 ran; 2 honoured the contract we drafted; 11 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · honoured contract
1ran · violated contract
7ran · our draft was wrong
1ran · fixture could not drive it
3ran
11unverified

Licence: 0 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; each sample names its own and says which. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

swish whwu95/Cap4Video/modules/until_module.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · 0f786c407fb1ee4c · report
check_attr whwu95/Cap4Video/modules/modeling_tv_titles_video.py official repository ran · violated contract MIT (permissive) · 47c2715feba5fd6b · report
compute_metrics whwu95/Cap4Video/metrics.py official repository ran fingerprinted MIT (permissive) · f45b80d057798fed · report
gelu whwu95/Cap4Video/modules/until_module.py official repository ran · honoured contract fingerprinted MIT (permissive) · 50e1ffed03f484ec · report
tensor_text_to_video_metrics whwu95/Cap4Video/metrics.py official repository ran MIT (permissive) · 66bf2e9b33bd27dd · report
update_attr whwu95/Cap4Video/modules/modeling_tv_titles_video.py official repository ran · our draft was wrong MIT (permissive) · d34a2226d2cf142e · report
url_to_filename whwu95/Cap4Video/modules/file_utils.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · af64ec220e8bcdbc · report
warmup_linear whwu95/Cap4Video/modules/optimization.py official repository ran · honoured contract fingerprinted MIT (permissive) · e0960afdb0d64aa7 · report
build_model whwu95/Cap4Video/modules/module_clip.py official repository unverified MIT (permissive) · 2fe63f02da32f330 · report
cached_path whwu95/Cap4Video/modules/file_utils.py official repository unverified MIT (permissive) · 349d780dd4a89a37 · report
download_model whwu95/Cap4Video/modules/module_clip.py official repository unverified MIT (permissive) · 1cd9a569d3043ec8 · report
filename_to_url whwu95/Cap4Video/modules/file_utils.py official repository unverified MIT (permissive) · db0ac56aaf6e35e6 · report
tensor_video_to_text_sim whwu95/Cap4Video/metrics.py official repository unverified MIT (permissive) · b6f9f3930836fe3f · report
warmup_constant whwu95/Cap4Video/modules/optimization.py official repository unverified MIT (permissive) · 59e4e730a9dc5dc6 · report
warmup_cosine whwu95/Cap4Video/modules/optimization.py official repository unverified MIT (permissive) · 20b3d25ba1b796e5 · report
basic_clean whwu95/BIKE/clip/simple_tokenizer.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · 98f385d847636a3e · report
drop_path whwu95/BIKE/clip/model.py community (archive-listed) ran · fixture could not drive it fingerprinted MIT (permissive) · 3ac6b7d76e8e3584 · report
get_pairs whwu95/BIKE/clip/simple_tokenizer.py community (archive-listed) ran · our draft was wrong MIT (permissive) · d919ae32e5e4e616 · report
multiple_samples_collate whwu95/BIKE/utils/Augmentation.py community (archive-listed) ran MIT (permissive) · 4927733091f71100 · report
update_dict whwu95/BIKE/train_attributes.py community (archive-listed) ran · our draft was wrong MIT (permissive) · d452c540ad7ecbbc · report
whitespace_clean whwu95/BIKE/clip/simple_tokenizer.py community (archive-listed) ran · our draft was wrong fingerprinted MIT (permissive) · 9542161e9640b858 · report
build_model whwu95/BIKE/clip/model.py community (archive-listed) unverified MIT (permissive) · 5d969c3a44340f51 · report
get_augmentation whwu95/BIKE/utils/Augmentation.py community (archive-listed) unverified MIT (permissive) · f38ab161348f5e80 · report
load whwu95/BIKE/clip/clip.py community (archive-listed) unverified MIT (permissive) · a34ba2785e08acbd · report
train_augmentation whwu95/BIKE/utils/Augmentation.py community (archive-listed) unverified MIT (permissive) · 8bd3efba1612299f · report

Tasks

Data AugmentationRetrievalVideo CaptioningVideo Retrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Retrieval DiDeMo Cap4Video text-to-video Mean Rank 10.5 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video text-to-video Median Rank 1 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video text-to-video R@1 52.0 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video text-to-video R@10 87.5 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video text-to-video R@5 79.4 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video video-to-text Mean Rank 7.3 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video video-to-text Median Rank 1 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video video-to-text R@1 51.2 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video video-to-text R@10 87.4 #23 of 40 Archive leaderboard report
Video Retrieval DiDeMo Cap4Video video-to-text R@5 78.5 #23 of 40 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video text-to-video Mean Rank 12.4 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video text-to-video Median Rank 1 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video text-to-video R@1 51.4 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video text-to-video R@10 83.9 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video text-to-video R@5 75.7 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video video-to-text Mean Rank 8.0 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video video-to-text Median Rank 2 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video video-to-text R@1 49.0 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video video-to-text R@10 85.0 #16 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA Cap4Video video-to-text R@5 75.2 #16 of 63 Archive leaderboard report
Video Retrieval MSVD Cap4Video text-to-video Mean Rank 8.3 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video text-to-video Median Rank 1 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video text-to-video R@1 51.8 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video text-to-video R@10 88.3 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video text-to-video R@5 80.8 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video video-to-text Mean Rank 2.4 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video video-to-text Median Rank 1 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video video-to-text R@1 70.0 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video video-to-text R@10 96.2 #10 of 24 Archive leaderboard report
Video Retrieval MSVD Cap4Video video-to-text R@5 93.2 #10 of 24 Archive leaderboard report
Video Retrieval VATEX Cap4Video text-to-video MeanR 2.7 #8 of 13 Archive leaderboard report
Video Retrieval VATEX Cap4Video text-to-video MedianR 1 #8 of 13 Archive leaderboard report
Video Retrieval VATEX Cap4Video text-to-video R@1 66.6 #8 of 13 Archive leaderboard report
Video Retrieval VATEX Cap4Video text-to-video R@10 97.0 #8 of 13 Archive leaderboard report
Video Retrieval VATEX Cap4Video text-to-video R@5 93.1 #8 of 13 Archive leaderboard report
Video Retrieval VATEX Cap4Video video-to-text R@1 80.9 #8 of 13 Archive leaderboard report
Video Retrieval VATEX Cap4Video video-to-text R@10 99.6 #8 of 13 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.

Methods

CLIP

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