Computer Vision News - May 2020

2 Summary Research 6 The high-level GAN network is shown in the following figure. The input and output are complex-valued images of the same size, where the real and imaginary components are considered as two separate channels. Here is the comparison between the Compressed Sensing reconstruction approach and the GANCS for a knee dataset (3T GE, 19 subjects, 3D FSE CUBE sequence). The main difference is that while the CS returns overly smooth images, the GANCS provides more detail but also slightly higher SSIM. The single-coil model is used. From a clinical MRI perspective, in an abdominal MRI and comparing the GANCS and CS-WV reconstruction, the former reveals tiny liver vessels, as shown in the image below. This shows the capability of providing detail in comparisonwith the Compressed Sensing reconstruction.