域適應-應用領域
當前基於深度學習的域適應,主要研究領域以及代表性的文章:
1.分類
Duplex Generative Adversarial Network for Unsupervised Domain Adaptation
Maximum Classifier Discrepancy for Unsupervised Domain Adaptation
2. 安防的行人重識別(Person Re-Identification
利用計算機視覺技術判斷影象或者視訊序列中是否存在特定行人的技術。廣泛被認為是一個影象檢索的子問題。給定一個監控行人影象,檢索跨裝置下的該行人影象。旨在彌補目前固定的攝像頭的視覺侷限,並可與行人檢測/行人跟蹤技術相結合,可廣泛應用於智慧視訊監控、智慧安保等領域。
Person Transfer GAN to Bridge Domain Gap for Person Re-Identification
3.自動駕駛背景下的語義分割( Semantic Segmentation)
Conditional Generative Adversarial Network for Structured Domain Adaptation
4.影象深度估計(Depth Estimation)
Single-Image Depth Estimation Based on Fourier Domain Analysis
Cross-Domain Self-Supervised Multi-Task Feature Learning Using Synthetic Imagery
AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimation
5. 目標檢測
Domain Adaptive Faster R-CNN for Object Detection in the Wild