DiMP:Learning Discriminative Model Prediction for Tracking
文章目录 摘要 背景 贡献 本文方法 工做1：提出一个新的损失函数并经过网络调参 工做2：提出一个强有力的迭代策略来确保模型快速收敛 参考 摘要 本文主要针对Siamese网络系列跟踪算法忽略目标背景信息的问题进行改进，旨在提升跟踪模型对于目标和背景的判别能力。具体地，本文设计了一个结合least-square regression和hinge loss优点的新的损失函数，并经过end-to-en
相关文章
- 1. DiMP:Learning Discriminative Model Prediction for Tracking
- 2. Learning Discriminative Model Prediction for Tracking --- 论文翻译
- 3. Learning Discriminative Model Prediction for Tracking 用于跟踪的学习判别模型预测
- 4. DIMP：Learning Discriminative Model Prediction for Tracking 学习判别模型预测的跟踪
- 5. Generative model & discriminative model
- 6. 论文笔记：Prediction-Tracking-Segmentation
- 7. Tracking算法：Discriminative Correlation Filter (DCF)
- 8. LC-RNN: A Deep Learning Model for Traffic Speed Prediction
- 9. A deep tree-based model for software defect prediction
- 10. Lecture4: Model-Free Prediction
- 更多相关文章...
- • Swift for 循环 - Swift 教程
- • Scala for循环 - Scala教程
- • 常用的分布式事务解决方案
- • PHP开发工具
相关标签/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。
欢迎关注本站公众号,获取更多信息
相关文章
- 1. DiMP:Learning Discriminative Model Prediction for Tracking
- 2. Learning Discriminative Model Prediction for Tracking --- 论文翻译
- 3. Learning Discriminative Model Prediction for Tracking 用于跟踪的学习判别模型预测
- 4. DIMP：Learning Discriminative Model Prediction for Tracking 学习判别模型预测的跟踪
- 5. Generative model & discriminative model
- 6. 论文笔记：Prediction-Tracking-Segmentation
- 7. Tracking算法：Discriminative Correlation Filter (DCF)
- 8. LC-RNN: A Deep Learning Model for Traffic Speed Prediction
- 9. A deep tree-based model for software defect prediction
- 10. Lecture4: Model-Free Prediction