NR-SIQA Using Convolutional Neural Network for Adaptive Feature Extraction
摘要： 在本文提出的算法中，使用CNN根據圖像的質量分數將圖像進行分類處理，得到網絡高層所提取到的視覺感知特徵。然後使用視覺顯著模型將提取的左右視圖特徵進行融合，同時多尺度下的視差圖統計特徵被提取出來，最終綜合以上特徵，使用SVR進行訓練。 方法詳述 首先，一個CNN模型從零開始訓練，自適應地提取圖像質量感知特徵。使用這個模型，從左右視圖中提取感知特徵，然後左右視圖基於顯著性權重進行融合得到融合特
相關文章
- 1. RCNN(Regions with Convolutional Neural Network Feature)
- 2. Fully Convolutional Network for Automatic Road Extraction from Satellite Imagery
- 3. Convolutional Neural Network - Object Detection
- 4. Distant Supervision for Relation Extraction via Piecewise Convolutional Neural Networks
- 5. 理解：Where to Focus: Query Adaptive Matching for Instance Retrieval Using Convolutional Feature Maps
- 6. Pre-Training Graph Neural Networks for Generic Structural Feature Extraction
- 7. Convolutional Neural Network
- 8. Convolutional Neural Network For Sentence Classification
- 9. Synaptic Strength For Convolutional Neural Network
- 10. Ensemble Neural Relation Extraction with Adaptive Boosting
- 更多相關文章...
- • Swift for 循環 - Swift 教程
- • Scala for循環 - Scala教程
- • RxJava操作符（六）Utility
- • Flink 數據傳輸及反壓詳解
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. RCNN(Regions with Convolutional Neural Network Feature)
- 2. Fully Convolutional Network for Automatic Road Extraction from Satellite Imagery
- 3. Convolutional Neural Network - Object Detection
- 4. Distant Supervision for Relation Extraction via Piecewise Convolutional Neural Networks
- 5. 理解：Where to Focus: Query Adaptive Matching for Instance Retrieval Using Convolutional Feature Maps
- 6. Pre-Training Graph Neural Networks for Generic Structural Feature Extraction
- 7. Convolutional Neural Network
- 8. Convolutional Neural Network For Sentence Classification
- 9. Synaptic Strength For Convolutional Neural Network
- 10. Ensemble Neural Relation Extraction with Adaptive Boosting