Abstract:Advanced Siamese visual object tracking architectures are jointly trained using pair-wise input images to perform target classification and bounding box regression. They have achieved promising results in recent benchmarks and competitions. However, the existing methods suffer from two limitations: First, though the Siamese structure can estimate the target state in an instance frame, provided the target appearance does not deviate too much from the template, the detection of the target in an image cannot be guaranteed in the presence of severe appearance variations. Second, despite the classification and regression tasks sharing the same output from the backbone network, their specific modules and loss functions are invariably designed independently, without promoting any interaction. Yet, in a general tracking task, the centre classification and bounding box regression tasks are collaboratively working to estimate the final target location. To address the above issues, it is essential to perform target-agnostic detection so as to promote cross-task interactions in a Siamese-based tracking framework. In this work, we endow a novel network with a target-agnostic object detection module to complement the direct target inference, and to avoid or minimise the misalignment of the key cues of potential template-instance matches. To unify the multi-task learning formulation, we develop a cross-task interaction module to ensure consistent supervision of the classification and regression branches, improving the synergy of different branches. To eliminate potential inconsistencies that may arise within a multi-task architecture, we assign adaptive labels, rather than fixed hard labels, to supervise the network training more effectively. The experimental results obtained on several benchmarks, i.e., OTB100, UAV123, VOT2018, VOT2019, and LaSOT, demonstrate the effectiveness of the advanced target detection module, as well as the cross-task interaction, exhibiting superior tra- king performance as compared with the state-of-the-art tracking methods.

Trajectory Guided Robust Visual Object Tracking with Selective Remedy.

Multi-features Guided Robust Visual Tracking.

Robust Object Tracking with Bidirectional Corner Matching and Trajectory Smoothness Algorithm.

Masked and Dynamic Siamese Network for Robust Visual Tracking.

SiamFC++: Towards Robust and Accurate Visual Tracking with Target Estimation Guidelines

Toward Robust Visual Object Tracking With Independent Target-Agnostic Detection and Effective Siamese Cross-Task Interaction

Visual Object Tracking Via Guessing and Matching

Siamese Network for Object Tracking with Multi-granularity Appearance Representations

SiamRDT: An Object Tracking Algorithm Based on a Reliable Dynamic Template

SiamMGT: robust RGBT tracking via graph attention and reliable modality weight learning

JROTM: Jointly Reinforced Object Tracking with Temporal Content Reference and Motion Guidance

Non-linear Target Trajectory Prediction for Robust Visual Tracking

Dynamic Attention Guided Multi-Trajectory Analysis for Single Object Tracking

Siamese Residual Network for Efficient Visual Tracking

Siamese Graph Attention Networks for Robust Visual Object Tracking.

Distillation, Ensemble and Selection for building a Better and Faster Siamese based Tracker

Local to Global Tracker: A Siamese Network for Long-term Tracking

Discriminative and Robust Online Learning for Siamese Visual Tracking

SiamRCR: Reciprocal Classification and Regression for Visual Object Tracking

Long-term object tracking based on joint tracking and detection strategy with Siamese network

Improved Fully Convolutional Siamese Networks for Visual Object Tracking Based on Response Behaviour Analysis.