DCFG: Diverse Cross-Channel Fine-Grained Feature Learning and Progressive Fusion Siamese Tracker for Thermal Infrared Target Tracking

2025-04-19

AI Summary

To address insufficient discriminative feature representation in thermal infrared (TIR) object tracking, this paper proposes a Siamese tracker based on cross-channel fine-grained feature learning and progressive fusion. The method introduces two key innovations: (1) a novel cross-channel fine-grained feature learning mechanism integrating mask-guided feature suppression, channel reordering and equalization, inter-layer combination, feature redirection, and channel shuffling; and (2) a cross-channel orthogonal loss function jointly optimizing feature diversity and discriminability. Evaluated on standard TIR benchmarks, the tracker achieves accuracy scores of 0.81 and 0.78 on VOT-TIR 2015 and 2017, respectively, and consistently outperforms state-of-the-art methods across all metrics on LSOTB-TIR and PTB-TIR. These results demonstrate the effectiveness and generalizability of the proposed framework for thermal imaging scenarios.

Technology Category

• Computer Vision: Motion & Tracking
• Search and Optimization: Learning to Search
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Tracking, profiling, and countermeasures against them
• Systems and Infrastructure for Web, Mobile and WoT: Experiences and lessons learnt from Web-based algorithms and system deployments

Abstract

To address the challenge of capturing highly discriminative features in ther-mal infrared (TIR) tracking, we propose a novel Siamese tracker based on cross-channel fine-grained feature learning and progressive fusion. First, we introduce a cross-channel fine-grained feature learning network that employs masks and suppression coefficients to suppress dominant target features, en-abling the tracker to capture more detailed and subtle information. The net-work employs a channel rearrangement mechanism to enhance efficient in-formation flow, coupled with channel equalization to reduce parameter count. Additionally, we incorporate layer-by-layer combination units for ef-fective feature extraction and fusion, thereby minimizing parameter redun-dancy and computational complexity. The network further employs feature redirection and channel shuffling strategies to better integrate fine-grained details. Second, we propose a specialized cross-channel fine-grained loss function designed to guide feature groups toward distinct discriminative re-gions of the target, thus improving overall target representation. This loss function includes an inter-channel loss term that promotes orthogonality be-tween channels, maximizing feature diversity and facilitating finer detail capture. Extensive experiments demonstrate that our proposed tracker achieves the highest accuracy, scoring 0.81 on the VOT-TIR 2015 and 0.78 on the VOT-TIR 2017 benchmark, while also outperforming other methods across all evaluation metrics on the LSOTB-TIR and PTB-TIR benchmarks.

Problem

• Enhancing discriminative feature capture in TIR tracking
• Reducing parameter redundancy and computational complexity
• Improving target representation via diverse feature learning

Innovation

• Cross-channel fine-grained feature learning with masks
• Progressive fusion with channel rearrangement and equalization
• Specialized loss function for diverse discriminative regions