We present the multi-timescale collaborative tracker for single object tracking. The tracker simultaneously utilizes different types of "forces", namely attraction, repulsion and support, to take advantage of their complementary strengths. We model the three forces via three components that are learned from the sample sets with different timescales. The long-term descriptive component attracts the target sample, while the medium-term discriminative component repulses the target from the background. They are collaborated in the appearance model to benefit each other. The short-term regressive component combines the votes of the auxiliary samples to predict the target's position, forming the context-aware motion model. The appearance model and the motion model collaboratively determine the target state, and the optimal state is estimated by a novel coarse-to-fine search strategy. We have conducted an extensive set of experiments on the standard 50 video benchmark. The results confirm the effectiveness of each component and their collaboration, outperforming current state-of-the-art methods.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1109/TPAMI.2016.2539956 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
A lightweight method of integrated local load forecasting and control of edge computing in active distribution networks.
iScience
August 2024
Key Laboratory of Smart Grid of Ministry of Education, Tianjin University, Tianjin 300072, China.
The strong resource constraints of edge-computing devices and the dynamic evolution of load characteristics put forward higher requirements for forecasting methods of active distribution networks. This paper proposes a lightweight adaptive ensemble learning method for local load forecasting and predictive control of active distribution networks based on edge computing in resource constrained scenarios. First, the adaptive sparse integration method is proposed to reduce the model scale.
View Article and Find Full Text PDFMulti-timescale variation characteristics of PM in different regions of China during 2014-2022.
Sci Total Environ
April 2024
State Key Laboratory of Pollution Control and Resources Reuse, School of Environment, Nanjing University, Nanjing 210023, China; Collaborative Innovation Center of Atmospheric Environment and Equipment Technology, Nanjing University of Information Science & Technology, Nanjing 210044, China. Electronic address:
Over the past decade, China has achieved a significant reduction in PM concentrations. Due to the diversity of natural and artificial factors, regional differences are remarkable in the variation characteristics and have not been well addressed in previous studies. Based on hourly observed PM concentrations from 2014 to 2022, this study conducted a comprehensive analysis of variation characteristics on annual, seasonal, and diurnal scales, with a special focus on differences across major regions.
View Article and Find Full Text PDFFew-Shot Learning in Spiking Neural Networks by Multi-Timescale Optimization.
Neural Comput
August 2021
College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China, and Zhejiang Lab, Hangzhou 311121, China
Learning new concepts rapidly from a few examples is an open issue in spike-based machine learning. This few-shot learning imposes substantial challenges to the current learning methodologies of spiking neuron networks (SNNs) due to the lack of task-related priori knowledge. The recent learning-to-learn (L2L) approach allows SNNs to acquire priori knowledge through example-level learning and task-level optimization.
View Article and Find Full Text PDFThe multi-timescale temporal patterns and dynamics of land surface temperature using Ensemble Empirical Mode Decomposition.
Sci Total Environ
February 2019
Faculty of Geo-Information Science and Earth Observation, University of Twente, Enschede 7500, the Netherlands. Electronic address:
Temporal variation patterns of Land Surface Temperature (LST) under different time scales are crucial in understanding the response of urban thermal environment to different forcings. However, there is no integrated toolset to extract such patterns from satellite remotely sensed time series LST (TSLST) data. This paper presents a workflow to extract the multi-timescale temporal patterns and dynamics from nonlinear and non-stationary TSLST data by taking Wuhan, China as case study.
View Article and Find Full Text PDFWe present the multi-timescale collaborative tracker for single object tracking. The tracker simultaneously utilizes different types of "forces", namely attraction, repulsion and support, to take advantage of their complementary strengths. We model the three forces via three components that are learned from the sample sets with different timescales.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!