A novel region adaptive learning-based super resolution algorithm for human face images is proposed, which divides a face image into flat regions and detailed regions.
提出一种基于区域自适应学习的人脸图像超分辨率复原算法。
The method considers that a high-resolution face image is composed of the global face image and local high-frequency details.
该方法将一幅高分辨率人脸图像视为由全局脸和局部高频细节两者组成的。
Finally, the global face image and local details are combined to the final face super-resolution results.
最后，将全局脸图像和局部细节信息相结合，得到最终的人脸超分辨率结果。
A new super-resolution face image reconstruction method based on linear object-class theory is proposed to deal with the problem of low-resolution face images.
为了解决单幅低分辨率人脸图像重构问题，提出了基于线性物体类理论重构超分辨率人脸图像的新方法。
Camera supports ZSL exposure, continuous focus, image zoom, capturing full-resolution snapshots while shooting videos, face detection and others.
相机支持ZSL曝光，连续对焦，图像缩放，捕捉全分辨率的快照，在拍摄影片、人脸识别和其他应用中都有了全新的体验。
With this method, the team used the device to scan and create a 3-D image of the "hills and valleys" on the front face of a U. S. penny -- with micron-level resolution -- from half ameter away.
借助此方法，设备在0.5米远的地方就能够清晰地扫描一美分硬币正面“错落有致的花纹（林肯头像）”，并创建其三维图像，而且分辨率达到微米级。
According to the characteristics of the morphable model, we refine an input image in the same way as the model, and then rebuild face model through multi-resolution matching.
根据形变模型的构造特点，把待匹配图像按照与模型相同方式进行加细，然后进行多分辨的人脸模型匹配。
According to the characteristics of the morphable model, we refine an input image in the same way as the model, and then rebuild face model through multi-resolution matching.
根据形变模型的构造特点，把待匹配图像按照与模型相同方式进行加细，然后进行多分辨的人脸模型匹配。
应用推荐