EML-GAN: Generative Adversarial Network-Based End-to-End Multi-Task Learning Architecture for Super-Resolution Reconstruction and Scene Classification of Low-Resolution Remote Sensing Imagery.
Weihuan DengQiqi ZhuXiongli SunWeihua LinQingfeng GuanPublished in: IGARSS (2021)
Keyphrases
- face recognition
- super resolution reconstruction
- low resolution
- high resolution
- super resolution
- low resolution images
- remote sensing
- space time
- high resolution images
- image super resolution
- feature extraction
- high quality
- object recognition
- natural scenes
- image representation
- multi frame
- image classification
- generative model
- visual words
- natural images
- image patches
- image quality
- sparse representation
- co occurrence
- motion estimation
- multiscale