paper-with-me

홈 › Papers

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

2025-03-24 · CVPR 2025 1 · 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.

📄 PDF Abstract BibTeX arXiv:2503.18338

Code (1)

wenruicai/spmtrack 공식 구현 pytorch

Tasks

Mixture-of-Expertsparameter-efficient fine-tuningRelationVisual Object TrackingVisual Tracking

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Multi-Head Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Residual Connection 설명 없음
ADOPT Please enter a description about the method here
Adam 설명 없음

Similar Papers 제목 키워드 기반

A Graph Prompt Fine-Tuning Method for WSN Spatio-Temporal Correlation Anomaly Detection

2026-01-19 · Miao Ye, Jing Cui, Yuan huang, Qian He 외 arxiv

Anomaly detection of multi-temporal modal data in Wireless Sensor Network (WSN) can provide an important guarantee for reliable network operation. Existing anomaly detection methods in multi-temporal modal data scenarios…

Self-Supervised LearningGraph Neural NetworkAnomaly Detection

ST-$π$: Structured SpatioTemporal VLA for Robotic Manipulation

2026-04-20 · Chuanhao Ma, Hanyu Zhou, Shihan Peng, Yan Li 외 arxiv

Vision-language-action (VLA) models have achieved great success on general robotic tasks, but still face challenges in fine-grained spatiotemporal manipulation. Typically, existing methods mainly embed spatiotemporal kno…

Expand and Compress: Exploring Tuning Principles for Continual Spatio-Temporal Graph Forecasting

2024-10-16 · Wei Chen, Yuxuan Liang

The widespread deployment of sensing devices leads to a surge in data for spatio-temporal forecasting applications such as traffic flow, air quality, and wind energy. Although spatio-temporal graph neural networks have a…

Graph Neural NetworkSpatio-Temporal Forecasting

Spatiotemporal Field Generation Based on Hybrid Mamba-Transformer with Physics-informed Fine-tuning

2025-05-16 · Peimian Du, Jiabin Liu, Xiaowei Jin, WangMeng Zuo 외

This research confronts the challenge of substantial physical equation discrepancies encountered in the generation of spatiotemporal physical fields through data-driven trained models. A spatiotemporal physical field gen…

MambaSelf-Supervised Learning

ST-Adapter: Parameter-Efficient Image-to-Video Transfer Learning

2022-06-27 · Junting Pan, Ziyi Lin, Xiatian Zhu, Jing Shao 외

Capitalizing on large pre-trained models for various downstream tasks of interest have recently emerged with promising performance. Due to the ever-growing model size, the standard full fine-tuning based task adaptation …

Action ClassificationAction Recognitionparameter-efficient fine-tuningTemporal Action Localization+2