MLT：Deep Meta Learning for Real-Time Target-Aware Visual Tracking
文章目錄 摘要 背景 貢獻 本文方法 摘要 本文針對傳統的基於深度卷積特徵的目標跟蹤算法爲了更好適應目標形變，須要進行復雜的優化/更新工做的問題進行改進，旨在保持較高跟蹤效果的同時實現跟蹤實時性。具體地，本文引入meta-learner網絡學習target-specific 特徵，與傳統的Siamese網絡提取的general特徵進行融合，獲得自適應的目標特徵，再進行後續的類似性計算。web 背景
相關文章
- 1. MLT：Deep Meta Learning for Real-Time Target-Aware Visual Tracking
- 2. CREST: Convolutional Residual Learning for Visual Tracking
- 3. Learning Spatially Regularized Correlation Filters for Visual Tracking （SRDCF）
- 4. Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- 5. Deep Reinforcement Learning with Iterative Shift for Visual Tracking
- 6. MDNet: Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- 7. Particle filter for visual tracking
- 8. simple online and realtime tracking
- 9. Tracking by Instance Detection: A Meta-Learning Approach
- 10. Multihypothesis Trajectory Analysis for Robust Visual Tracking
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • Java Agent入門實戰（三）-JVM Attach原理與使用
- • Java Agent入門實戰（一）-Instrumentation介紹與使用
相關標籤/搜索
每日一句
-
每一个你不满意的现在，都有一个你没有努力的曾经。
歡迎關注本站公眾號,獲取更多信息
相關文章
- 1. MLT：Deep Meta Learning for Real-Time Target-Aware Visual Tracking
- 2. CREST: Convolutional Residual Learning for Visual Tracking
- 3. Learning Spatially Regularized Correlation Filters for Visual Tracking （SRDCF）
- 4. Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- 5. Deep Reinforcement Learning with Iterative Shift for Visual Tracking
- 6. MDNet: Learning Multi-Domain Convolutional Neural Networks for Visual Tracking
- 7. Particle filter for visual tracking
- 8. simple online and realtime tracking
- 9. Tracking by Instance Detection: A Meta-Learning Approach
- 10. Multihypothesis Trajectory Analysis for Robust Visual Tracking