Abstract:Hyperspectral videos can provide the spatial, spectral, and motion information of targets, which makes it possible to track camouflaged targets that are similar to the background. However, hyperspectral object tracking is a challenging task, due to the huge hyperspectral video data dimension and the “data hungry” problem for the model training. Insufficient training data can seriously interfere with the accuracy and generalization of the tracking models. In this paper, a dual deep Siamese network framework for hyperspectral object tracking (SiamHYPER) is proposed for learning a hyperspectral tracker from a pretrained RGB tracker in the case of the “data hungry” problem. Specifically, in addition to a pretrained RGB-based Siamese tracker, a hyperspectral target-aware module is designed to mine the spectral information during the target prediction, and a spatial-spectral cross-attention module is introduced to further fuse the deep spatial and spectral features extracted from the RGB tracker and the hyperspectral target-aware module. Benefiting from the guidance training of the RGB tracker, a robust hyperspectral object tracker can be trained effectively with only a small number of hyperspectral video samples, to overcome the “data hungry” problem. In the experiments conducted in this study, the SiamHYPER framework was verified using SiamBAN and SiamRPN++, with 13 000 frames of hyperspectral videos for training, and achieved the best performance on the publicly available hyperspectral dataset released as part of the WHISPERS Hyperspectral Object Tracking Challenge. The area under the curve (AUC) of SiamHYPER was increased by nearly 8.9% and 7.2%, respectively, when compared with the current state-of-the-art RGB-based and hyperspectral trackers. In addition, the processing speed of SiamHYPER was 19 FPS, which is much higher than that of the current state-of-the-art hyperspectral trackers. The source code is available at zhenliuzhenqi/HOT: Hyperspectral object tracking (github.com).

HSPTrack: Hyperspectral Sequence Prediction Tracker with Transformers

SiamHYPER: Learning a Hyperspectral Object Tracker from an RGB-Based Tracker

A Spectral–Spatial Transformer Fusion Method for Hyperspectral Video Tracking

Spectral-Spatial-Aware Transformer Fusion Network For Hyperspectral Object Tracking

Unsupervised Deep Hyperspectral Video Target Tracking and High Spectral-Spatial-Temporal Resolution (H³ Benchmark Dataset

An Anchor-Free Siamese Target Tracking Network for Hyperspectral Video.

A Transformer-Based Network for Hyperspectral Object Tracking.

SSTtrack: A Unified Hyperspectral Video Tracking Framework via Modeling Spectral-Spatial-Temporal Conditions

HHTrack: Hyperspectral Object Tracking Using Hybrid Attention

Transformer-Based Band Regrouping With Feature Refinement for Hyperspectral Object Tracking

Spectral-Spatial-Temporal Attention Network for Hyperspectral Tracking.

Multi-Band Hyperspectral Object Tracking: Leveraging Spectral Information Prompts and Spectral Scale-Aware Representation

Learning a Deep Ensemble Network with Band Importance for Hyperspectral Object Tracking.

A Fast Hyperspectral Tracking Method Via Channel Selection

Domain Adaptation-aware Transformer for Hyperspectral Object Tracking

A Fast Hyperspectral Object Tracking Method Based on Channel Selection Strategy

A Transformer-Based Three-Branch Siamese Network for Hyperspectral Object Tracking

PHTrack: Prompting for Hyperspectral Video Tracking

Hy-Tracker: A Novel Framework for Enhancing Efficiency and Accuracy of Object Tracking in Hyperspectral Videos

A Siamese Network-Based Tracking Framework for Hyperspectral Video

Object Tracking in Hyperspectral-Oriented Video with Fast Spatial-Spectral Features