LiteFlowNet: A Lightweight Convolutional Neural Network for Optical Flow Estimation
Abstract flownet效果好，但是需要160M的參數。創新點：1.使得前向傳播預測光流更爲效率通過在每一個金字塔層添加一個串聯網絡。2.添加一個novel flow regularization layer來改善異常值和模糊邊界的情況，這個層是通過使用feature-driven local convolution來實現的。3.我們的網絡擁有一個有效的金字塔特徵提取結構，並採用fea
相關文章
- 1. Optical Flow 發展歷程 (1)
- 2. PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes—2017（筆記）
- 3. FlowNet: Learning Optical Flow with Convolutional Networks
- 4. 論文解讀1-LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow Estimation
- 5. Tensorflow - Implement for a Convolutional Neural Network on MNIST.
- 6. A Cascaded Convolutional Neural Network for Single Image Dehazing
- 7. 《A Convolutional Neural Network Cascade for Face Detection》
- 8. A mixed-scale dense convolutional neural network for image analysis
- 9. Deep Alignment Network: A convolutional neural network for robust face alignment
- 10. FlowNet 2.0 Evolution of Optical Flow Estimation with Deep Networks
- 更多相關文章...
- • XSL-FO flow 對象 - XSL-FO 教程
- • Scala for循環 - Scala教程
- • Flink 數據傳輸及反壓詳解
- • YAML 入門教程
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. Optical Flow 發展歷程 (1)
- 2. PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes—2017（筆記）
- 3. FlowNet: Learning Optical Flow with Convolutional Networks
- 4. 論文解讀1-LiteFlowNet3: Resolving Correspondence Ambiguity for More Accurate Optical Flow Estimation
- 5. Tensorflow - Implement for a Convolutional Neural Network on MNIST.
- 6. A Cascaded Convolutional Neural Network for Single Image Dehazing
- 7. 《A Convolutional Neural Network Cascade for Face Detection》
- 8. A mixed-scale dense convolutional neural network for image analysis
- 9. Deep Alignment Network: A convolutional neural network for robust face alignment
- 10. FlowNet 2.0 Evolution of Optical Flow Estimation with Deep Networks