Related references
Note: Only part of the references are listed.
Article
Computer Science, Artificial Intelligence
Wei Liang et al.
Summary: In this paper, an improved Siamese tracking network, named GSiamMS, is proposed to address the challenges of large-scale changes and similar objects in visual tracking. By integrating Res2Net blocks and transformer modules, the developed method achieves superior performance compared to state-of-the-art methods, running at a speed of 38fps as demonstrated in extensive experiments on various visual tracking benchmarks.
KNOWLEDGE-BASED SYSTEMS
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Yutao Cui et al.
Summary: This paper presents a compact tracking framework called MixFormer, which utilizes mixed attention modules for simultaneous feature extraction and target information integration. The framework achieves state-of-the-art performance on five tracking benchmarks, demonstrating the effectiveness of the proposed approach.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Article
Engineering, Electrical & Electronic
Xiao Wang et al.
Summary: The study proposes a dynamic attention-guided multi-trajectory tracking strategy to improve the performance of single object trackers, maintaining diversified tracking results by constructing multiple target templates and introducing different attentions, and selecting the best trajectory to enhance tracking performance.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
(2021)
Article
Computer Science, Artificial Intelligence
Lianghua Huang et al.
Summary: GOT-10k is a large tracking database that covers over 560 classes of moving objects and 87 motion patterns, providing a unified training and evaluation platform for tracker development. Additionally, the introduction of a one-shot protocol for tracker evaluation avoids biased results towards familiar objects and promotes generalization in tracker development.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Computer Science, Artificial Intelligence
Fei Zhao et al.
Summary: Siamese networks are commonly used in visual tracking due to efficient localization, but they struggle to capture target appearance variations. This paper proposes a template updating method using reinforcement learning to dynamically update templates and network parameters in each frame, adapting to target appearance changes. Experimental results demonstrate the effectiveness of the proposed approach against state-of-the-art methods.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Xin Li et al.
Article
Computer Science, Artificial Intelligence
Zhiyuan Liang et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2020)
Article
Computer Science, Artificial Intelligence
Xin Li et al.
KNOWLEDGE-BASED SYSTEMS
(2019)
Article
Computer Science, Artificial Intelligence
Pengpeng Liang et al.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2015)