The Fixed Server Locality Gap of Count Load Assignment Games
研究通过边际贡献定价解决服务器分配问题，分析了不同工作负载和成本下的纳什均衡效率，并证明了在固定服务器数量下价格无序度的上界。
0 citationsRead paper
Selected work
Representative Papers
DCFG: Diverse Cross-Channel Fine-Grained Feature Learning and Progressive Fusion Siamese Tracker for Thermal Infrared Target Tracking
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.
0 citationsRead paper
FGSGT: Saliency-Guided Siamese Network Tracker Based on Key Fine-Grained Feature Information for Thermal Infrared Target Tracking
Thermal infrared (TIR) images suffer from low contrast and impoverished texture details, leading to weak discriminability, susceptibility to interference, and drift in visual trackers. To address these challenges, we propose SG-Siam—a fully end-to-end saliency-guided Siamese network. It employs fine-grained feature parallel learning convolutional blocks to extract discriminative local features; introduces a multi-level bilinear fusion module to enable cross-scale interaction among fine-grained features; and incorporates a residual refinement mechanism for saliency prediction jointly optimized with a saliency-constrained loss function, enabling attention-driven robust tracking. Extensive experiments demonstrate state-of-the-art performance: SG-Siam achieves top accuracy and success rates on PTB-TIR and LSOTB-TIR benchmarks. On VOT-TIR 2015 and 2017, it attains accuracy scores of 0.78 and 0.75, respectively—significantly outperforming existing methods.
0 citationsRead paper
Recent publications
Latest Papers
The Fixed Server Locality Gap of Count Load Assignment Games
研究通过边际贡献定价解决服务器分配问题，分析了不同工作负载和成本下的纳什均衡效率，并证明了在固定服务器数量下价格无序度的上界。
0 citationsRead paper
DCFG: Diverse Cross-Channel Fine-Grained Feature Learning and Progressive Fusion Siamese Tracker for Thermal Infrared Target Tracking
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.
0 citationsRead paper
FGSGT: Saliency-Guided Siamese Network Tracker Based on Key Fine-Grained Feature Information for Thermal Infrared Target Tracking
Thermal infrared (TIR) images suffer from low contrast and impoverished texture details, leading to weak discriminability, susceptibility to interference, and drift in visual trackers. To address these challenges, we propose SG-Siam—a fully end-to-end saliency-guided Siamese network. It employs fine-grained feature parallel learning convolutional blocks to extract discriminative local features; introduces a multi-level bilinear fusion module to enable cross-scale interaction among fine-grained features; and incorporates a residual refinement mechanism for saliency prediction jointly optimized with a saliency-constrained loss function, enabling attention-driven robust tracking. Extensive experiments demonstrate state-of-the-art performance: SG-Siam achieves top accuracy and success rates on PTB-TIR and LSOTB-TIR benchmarks. On VOT-TIR 2015 and 2017, it attains accuracy scores of 0.78 and 0.75, respectively—significantly outperforming existing methods.
0 citationsRead paper