Abstract:Siamese networks have achieved great success in object tracking due to the balance of precision and speed. However, Siamese trackers usually utilize the local feature of the last layer, which may degrade tracking performance in some difficult scenarios. In this paper, we propose a novel Channel and Spatial Attention-guided Residual learning framework for Tracking, referred to as CSART, which can improve feature representation of Siamese networks by exploiting self-attention mechanism to capture powerful contextual information. Specifically, to be efficient and seamless integration, different kinds of self-attention are appended on the template and search branches of Siamese networks respectively, that model global semantic inter-dependencies in channel and spatial dimensions. To avoid representation degradation, we consider to adaptively aggregate basic feature and its attention-weighted features with residual learning. Furthermore, a joint loss consisting of classic logistic loss as well as focal softmax loss is designed to emphasize difficult samples and guide the learning process of the whole model. Benefiting from the above scheme, CSART alleviates the over-fitting problem to some extent and enhances the discriminability. Extensive experiments on six popular tracking datasets indicate that the proposed tracker achieves better performance with a speed of 65 fps than other state-of-the-art trackers. (c) 2020 Elsevier B.V. All rights reserved. Siamese networks have achieved great success in object tracking due to the balance of precision and speed. However, Siamese trackers usually utilize the local feature of the last layer, which may degrade tracking performance in some difficult scenarios. In this paper, we propose a novel Channel and Spatial Attention-guided Residual learning framework for Tracking, referred to as CSART, which can improve feature representation of Siamese networks by exploiting self-attention mechanism to capture powerful contextual information. Specifically, to be efficient and seamless integration, different kinds of self-attention are appended on the template and search branches of Siamese networks respectively, that model global semantic inter-dependencies in channel and spatial dimensions. To avoid representation degradation, we consider to adaptively aggregate basic feature and its attention-weighted features with residual learning. Furthermore, a joint loss consisting of classic logistic loss as well as focal softmax loss is designed to emphasize difficult samples and guide the learning process of the whole model. Benefiting from the above scheme, CSART alleviates the over-fitting problem to some extent and enhances the discriminability. Extensive experiments on six popular tracking datasets indicate that the proposed tracker achieves better performance with a speed of 65 fps than other state-of-the-art trackers.

Trustworthy Dynamic Object Tracking Using Deep Reinforcement Learning with the Self-Attention Mechanism

Exploit Spatiotemporal Contextual Information for 3D Single Object Tracking Via Memory Networks

Beyond Traditional Driving Scenes: A Robotic-Centric Paradigm for 2D+3D Human Tracking Using Siamese Transformer Network

Reinforcement-Tracking: An End-to-End Trajectory Tracking Method Based on Self-Attention Mechanism

End-to-end Active Object Tracking Via Reinforcement Learning

Chained-Tracker: Chaining Paired Attentive Regression Results for End-to-End Joint Multiple-Object Detection and Tracking

Dynamic Object Tracking for Self-Driving Cars Using Monocular Camera and LIDAR.

End-to-End Active Object Tracking and Its Real-World Deployment Via Reinforcement Learning

On Deep Recurrent Reinforcement Learning for Active Visual Tracking of Space Noncooperative Objects

CSART: Channel and Spatial Attention-Guided Residual Learning for Real-Time Object Tracking.

Collaborative Deep Reinforcement Learning for Multi-object Tracking

Visual Tracking Via Hierarchical Deep Reinforcement Learning

Attention-Driven Memory Network for Online Visual Tracking.

Learning Attention for Object Tracking with Adversarial Learning Network

An Online and Flexible Multi-object Tracking Framework Using Long Short-Term Memory

Learning reinforced attentional representation for end-to-end visual tracking

Video Key Object Detection Network via Reinforcement Learning

Reinforcement-Tracking: An Effective Trajectory Tracking and Navigation Method for Autonomous Urban Driving

Deep Reinforcement Learning-Based End-to-End Control for UAV Dynamic Target Tracking

An IMM-Enabled Adaptive 3D Multi-Object Tracker for Autonomous Driving

Deep Reinforcement Learning-Based DQN Agent Algorithm for Visual Object Tracking in a Virtual Environmental Simulation