Papers › CenterCLIP: Token Clustering for Efficient Text-Video Retrieval

CenterCLIP: Token Clustering for Efficient Text-Video Retrieval

2 May 2022arXiv:2205.00823archive 2025-07-28

Shuai Zhao, Linchao Zhu, Xiaohan Wang, Yi Yang

Recently, large-scale pre-training methods like CLIP have made great progress in multi-modal research such as text-video retrieval. In CLIP, transformers are vital for modeling complex multi-modal relations. However, in the vision transformer of CLIP, the essential visual tokenization process, which produces discrete visual token sequences, generates many homogeneous tokens due to the redundancy nature of consecutive and similar frames in videos. This significantly increases computation costs and hinders the deployment of video retrieval models in web applications. In this paper, to reduce the number of redundant video tokens, we design a multi-segment token clustering algorithm to find the most representative tokens and drop the non-essential ones. As the frame redundancy occurs mostly in consecutive frames, we divide videos into multiple segments and conduct segment-level clustering. Center tokens from each segment are later concatenated into a new sequence, while their original spatial-temporal relations are well maintained. We instantiate two clustering algorithms to efficiently find deterministic medoids and iteratively partition groups in high dimensional space. Through this token clustering and center selection procedure, we successfully reduce computation costs by removing redundant visual tokens. This method further enhances segment-level semantic alignment between video and text representations, enforcing the spatio-temporal interactions of tokens from within-segment frames. Our method, coined as CenterCLIP, surpasses existing state-of-the-art by a large margin on typical text-video benchmarks, while reducing the training memory cost by 35% and accelerating the inference speed by 14% at the best case. The code is available at https://github.com/mzhaoshuai/CenterCLIP.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

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

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 1 code sample harvested from 1 repository linked to this paper: 0 ran · 1 unverified.

By repository: official repository: 1 sample from 1 repository, 0 ran (0 with no instrument failure; 0 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

mzhaoshuai/CenterCLIP officialmentioned on GitHubpytorchSyntology: 1 sample, 0 ran · 1 unverified 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

1 sample harvested: 0 ran · 1 unverified. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1unverified

Licence: 1 of the 1 sample is 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 mzhaoshuai/CenterCLIP. “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 “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (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 are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. 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 six 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; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. 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.

mzhaoshuai/CenterCLIP official repository 1 sample, 0 ran · 1 unverified
get_cluster_inter modules/cluster/cluster.py unverified licence not identified · pointer only · 93f1c0cd5419e51b · report

Tasks

ClusteringRetrievalVideo Retrieval

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) text-to-video Mean Rank 5.7 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) text-to-video Median Rank 2 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) text-to-video R@1 46.2 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) text-to-video R@10 87.6 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) text-to-video R@5 77.0 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) video-to-text Mean Rank 5.5 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) video-to-text Median Rank 2 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) video-to-text R@1 46.7 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) video-to-text R@10 88.0 #20 of 31 Archive leaderboard report
Video Retrieval ActivityNet CenterCLIP (ViT-B/16) video-to-text R@5 77.1 #20 of 31 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) text-to-video Mean Rank 47.3 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) text-to-video Median Rank 8 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) text-to-video R@1 24.2 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) text-to-video R@10 55.9 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) text-to-video R@5 46.2 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) video-to-text Mean Rank 41.3 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) video-to-text Median Rank 7 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) video-to-text R@1 24.5 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) video-to-text R@10 55.8 #20 of 38 Archive leaderboard report
Video Retrieval LSMDC CenterCLIP (ViT-B/16) video-to-text R@5 46.4 #20 of 38 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) text-to-video Mean Rank 13.8 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) text-to-video Median Rank 2 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) text-to-video R@1 48.4 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) text-to-video R@10 82.0 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) text-to-video R@5 73.8 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) video-to-text Mean Rank 10.2 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) video-to-text Median Rank 2 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) video-to-text R@1 47.7 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) video-to-text R@10 83.3 #26 of 63 Archive leaderboard report
Video Retrieval MSR-VTT-1kA CenterCLIP (ViT-B/16) video-to-text R@5 75.0 #26 of 63 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) text-to-video Mean Rank 8.4 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) text-to-video Median Rank 1 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) text-to-video R@1 50.6 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) text-to-video R@10 88.4 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) text-to-video R@5 80.3 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) video-to-text Mean Rank 3.0 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) video-to-text Median Rank 1 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) video-to-text R@1 68.4 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) video-to-text R@10 95.0 #11 of 24 Archive leaderboard report
Video Retrieval MSVD CenterCLIP (ViT-B/16) video-to-text R@5 90.1 #11 of 24 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

AttentionCLIPDense ConnectionsLayer NormalizationLinear LayerMulti-Head AttentionResidual ConnectionSPEEDSoftmaxVision Transformer

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