Papers › Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance

Tracking Meets LoRA: Faster Training, Larger Model, Stronger Performance

8 Mar 2024arXiv:2403.05231archive 2025-07-28

Liting Lin, Heng Fan, Zhipeng Zhang, YaoWei Wang, Yong Xu, Haibin Ling

Motivated by the Parameter-Efficient Fine-Tuning (PEFT) in large language models, we propose LoRAT, a method that unveils the power of large ViT model for tracking within laboratory-level resources. The essence of our work lies in adapting LoRA, a technique that fine-tunes a small subset of model parameters without adding inference latency, to the domain of visual tracking. However, unique challenges and potential domain gaps make this transfer not as easy as the first intuition. Firstly, a transformer-based tracker constructs unshared position embedding for template and search image. This poses a challenge for the transfer of LoRA, usually requiring consistency in the design when applied to the pre-trained backbone, to downstream tasks. Secondly, the inductive bias inherent in convolutional heads diminishes the effectiveness of parameter-efficient fine-tuning in tracking models. To overcome these limitations, we first decouple the position embeddings in transformer-based trackers into shared spatial ones and independent type ones. The shared embeddings, which describe the absolute coordinates of multi-resolution images (namely, the template and search images), are inherited from the pre-trained backbones. In contrast, the independent embeddings indicate the sources of each token and are learned from scratch. Furthermore, we design an anchor-free head solely based on MLP to adapt PETR, enabling better performance with less computational overhead. With our design, 1) it becomes practical to train trackers with the ViT-g backbone on GPUs with only memory of 25.8GB (batch size of 16); 2) we reduce the training time of the L-224 variant from 35.0 to 10.8 GPU hours; 3) we improve the LaSOT SUC score from 0.703 to 0.742 with the L-224 variant; 4) we fast the inference speed of the L-224 variant from 52 to 119 FPS. Code and models are available at https://github.com/LitingLin/LoRAT.

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 5 code samples harvested from 1 repository linked to this paper: 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified. Where Syntology's instrument failed: 2 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 5 samples from 1 repository, 3 ran (1 with no instrument failure; 2 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.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

litinglin/lorat officialmentioned in papermentioned on GitHubpytorchApache-2.0Syntology: 5 samples, 3 ran (2 instrument-failed) · 2 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

5 samples harvested: 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 2 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · our draft was wrong
1ran
2unverified

Licence: 0 of the 5 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 litinglin/lorat. “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. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “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.

litinglin/lorat official repository 5 samples, 3 ran (2 instrument-failed) · 2 unverified
find_all_frozen_nn_linear_names trackit/models/methods/LoRAT/funcs/vit_lora_utils.py ran · our draft was wrong Apache-2.0 (permissive) · 5a695ef0811cab18 · report fixture: signature model:callable/0/? · returned list · level recorded 2026-09-24
freeze_vit_backbone_ trackit/models/methods/LoRAT/funcs/vit_backbone_freeze.py ran · our draft was wrong Apache-2.0 (permissive) · 4b916f2f46e88d85 · report fixture: signature model:callable/0/?, freeze_out_norm:bool/0/bool · returned MinimalViT · level recorded 2026-09-24
generate_LoRAT_sample_data trackit/models/methods/LoRAT/funcs/sample_data.py ran Apache-2.0 (permissive) · c311b3c84e280fa0 · report fixture: signature z_feat_size:tuple/1/int, x_feat_size:tuple/1/int, patch_size:tuple/1/int, batch_size:int/0/int, device:device/0/?, dtype:dtype/0/? · returned tuple · level recorded 2026-09-24
enable_lora_ trackit/models/methods/LoRAT/funcs/vit_lora_utils.py unverified Apache-2.0 (permissive) · 7dcc0654aaf67867 · report
state_dict_with_lora_meta_attributes trackit/models/methods/LoRAT/funcs/vit_lora_utils.py unverified Apache-2.0 (permissive) · f6f3949d8af0fe7f · report

Tasks

Visual 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 LoRAT-g-378 Average Overlap 78.9 #8 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k LoRAT-g-378 Success Rate 0.5 87.8 #8 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k LoRAT-g-378 Success Rate 0.75 80.7 #8 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k LoRAT-L-378 Average Overlap 77.5 #13 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k LoRAT-L-378 Success Rate 0.5 86.2 #13 of 42 Archive leaderboard report
Visual Object Tracking GOT-10k LoRAT-L-378 Success Rate 0.75 78.1 #13 of 42 Archive leaderboard report
Visual Object Tracking LaSOT LoRAT-g-378 AUC 76.2 #4 of 46 Archive leaderboard report
Visual Object Tracking LaSOT LoRAT-g-378 Normalized Precision 85.3 #4 of 46 Archive leaderboard report
Visual Object Tracking LaSOT LoRAT-g-378 Precision 83.5 #4 of 46 Archive leaderboard report
Visual Object Tracking LaSOT LoRAT-L-378 AUC 75.1 #6 of 46 Archive leaderboard report
Visual Object Tracking LaSOT LoRAT-L-378 Normalized Precision 84.1 #6 of 46 Archive leaderboard report
Visual Object Tracking LaSOT LoRAT-L-378 Precision 82.0 #6 of 46 Archive leaderboard report
Visual Object Tracking LaSOT-ext LoRAT-L-378 AUC 56.6 #3 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext LoRAT-L-378 Normalized Precision 69.0 #3 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext LoRAT-L-378 Precision 65.1 #3 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext LoRAT-g-378 AUC 56.5 #4 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext LoRAT-g-378 Normalized Precision 69.0 #4 of 18 Archive leaderboard report
Visual Object Tracking LaSOT-ext LoRAT-g-378 Precision 64.9 #4 of 18 Archive leaderboard report
Visual Object Tracking NeedForSpeed LoRAT-g-378 AUC 0.681 #5 of 10 Archive leaderboard report
Visual Object Tracking NeedForSpeed LoRAT-L-378 AUC 0.667 #8 of 10 Archive leaderboard report
Visual Object Tracking TNL2K LoRAT-g-378 AUC 62.7 #5 of 16 Archive leaderboard report
Visual Object Tracking TNL2K LoRAT-g-378 precision 67.8 #5 of 16 Archive leaderboard report
Visual Object Tracking TNL2K LoRAT-L-378 AUC 62.3 #6 of 16 Archive leaderboard report
Visual Object Tracking TNL2K LoRAT-L-378 precision 67.0 #6 of 16 Archive leaderboard report
Visual Object Tracking TrackingNet LoRAT-g-378 Accuracy 86.0 #9 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet LoRAT-g-378 Normalized Precision 90.2 #9 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet LoRAT-g-378 Precision 86.1 #9 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet LoRAT-L-378 Accuracy 85.6 #10 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet LoRAT-L-378 Normalized Precision 89.7 #10 of 40 Archive leaderboard report
Visual Object Tracking TrackingNet LoRAT-L-378 Precision 85.4 #10 of 40 Archive leaderboard report
Visual Object Tracking UAV123 LoRAT-g-378 AUC 0.739 #1 of 16 Archive leaderboard report
Visual Object Tracking UAV123 LoRAT-L-378 AUC 0.725 #3 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.

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