Papers › SPMTrack: Spatio-Temporal Parameter-Efficient Fine-Tuning with Mixture of Experts for...

SPMTrack: Spatio-Temporal Parameter-Efficient Fine-Tuning with Mixture of Experts for Scalable Visual Tracking

24 Mar 2025CVPR 2025 1arXiv:2503.18338archive 2025-07-28

Wenrui Cai, Qingjie Liu, Yunhong Wang

Most state-of-the-art trackers adopt one-stream paradigm, using a single Vision Transformer for joint feature extraction and relation modeling of template and search region images. However, relation modeling between different image patches exhibits significant variations. For instance, background regions dominated by target-irrelevant information require reduced attention allocation, while foreground, particularly boundary areas, need to be be emphasized. A single model may not effectively handle all kinds of relation modeling simultaneously. In this paper, we propose a novel tracker called SPMTrack based on mixture-of-experts tailored for visual tracking task (TMoE), combining the capability of multiple experts to handle diverse relation modeling more flexibly. Benefiting from TMoE, we extend relation modeling from image pairs to spatio-temporal context, further improving tracking accuracy with minimal increase in model parameters. Moreover, we employ TMoE as a parameter-efficient fine-tuning method, substantially reducing trainable parameters, which enables us to train SPMTrack of varying scales efficiently and preserve the generalization ability of pretrained models to achieve superior performance. We conduct experiments on seven datasets, and experimental results demonstrate that our method significantly outperforms current state-of-the-art trackers. The source code is available at https://github.com/WenRuiCai/SPMTrack.

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="2503.18338")

Code

Syntology Ran 0 of 1 code samples harvested from 1 repository linked to this paper; 1 has no recorded run.

By repository: official repository: 1 sample from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

wenruicai/spmtrack officialmentioned in papermentioned on GitHubpytorchApache-2.0 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; 0 honoured the contract we drafted; 1 has no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1unverified

Licence: 0 of the 1 sample 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 WenRuiCai/SPMTrack. “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.

drop_add_residual_stochastic_depth WenRuiCai/SPMTrack/trackit/models/methods/SPMTrack/modules/mixformer_attn.py official repository unverified Apache-2.0 (permissive) · 29d3c222b6dd6659 · report

Tasks

Mixture-of-ExpertsVisual Object TrackingVisual Trackingparameter-efficient fine-tuning

1 archive task tag without a task page not shown.

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Visual Object Tracking GOT-10k SPMTrack-G Average Overlap 81 #3 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-G Success Rate 0.5 89.2 #3 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-G Success Rate 0.75 82.3 #3 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-L Average Overlap 80 #5 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-L Success Rate 0.5 89.4 #5 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-L Success Rate 0.75 79.9 #5 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-B Average Overlap 76.5 #17 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-B Success Rate 0.5 85.9 #17 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k SPMTrack-B Success Rate 0.75 76.3 #17 of 42 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-G AUC 77.4 #1 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-G Normalized Precision 86.6 #1 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-G Precision 85 #1 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-L AUC 76.8 #2 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-L Normalized Precision 85.9 #2 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-L Precision 84 #2 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-B AUC 74.9 #8 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-B Normalized Precision 84 #8 of 46 Archive leaderboard report
Visual Object Tracking LaSOT SPMTrack-B Precision 81.7 #8 of 46 Archive leaderboard report
Visual Object Tracking NeedForSpeed SPMTrack-B AUC 0.674 #7 of 10 Archive leaderboard report
Visual Object Tracking OTB-2015 SPMTrack-B AUC 0.727 #1 of 18 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-G AUC 64.7 #2 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-G Normalized Precision 82.6 #2 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-G precision 70.6 #2 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-L AUC 63.7 #3 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-L Normalized Precision 81.5 #3 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-L precision 69.2 #3 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-B AUC 62.0 #7 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-B Normalized Precision 79.7 #7 of 16 Archive leaderboard report
Visual Object Tracking TNL2K SPMTrack-B precision 66.7 #7 of 16 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-G Accuracy 87.3 #2 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-G Normalized Precision 91.4 #2 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-G Precision 88.1 #2 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-L Accuracy 86.9 #3 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-L Normalized Precision 91 #3 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-L Precision 87.2 #3 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-B Accuracy 86.1 #7 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-B Normalized Precision 90.2 #7 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet SPMTrack-B Precision 85.6 #7 of 40 Archive leaderboard report
Visual Object Tracking UAV123 SPMTrack-B AUC 0.717 #5 of 16 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

ADOPTAbsolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutLabel SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformerVision 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