This Week in 3D Imaging

|
|
0 comments

Research:

3D ImagingMedMNIST v2, a sizable collection of standardized biological image datasets similar to MNIST that includes 12 datasets for 2D and 6 datasets for 3D, was unveiled by researchers from Shanghai Jiao Tong University. The generated dataset, which consists of 708,069 2D photos and 9,998 3D images overall, could serve a variety of research / teaching goals in biological image analysis, computer vision, and machine learning. They evaluated a number of benchmark techniques on MedMNIST v2, including 2D and 3D neural networks and open-source and for-profit AutoML systems.

3D ImagingWithout collecting 3D face scans, IEEE researchers suggested a novel method for learning a nonlinear 3DMM model from a vast collection of in-the-wild face photos. A network encoder specifically estimates the projection, lighting, shape, and albedo characteristics given a face image as input. End-to-end training of the entire network is possible with minimal monitoring. They showed how their nonlinear 3DMM's contribution to face editing, 3D reconstruction, and face alignment outperformed its linear version in terms of representation power.

Open Source News:

Open Source NewsThe general release of StarlingX 8.0, which offers telcos a new open source cloud and edge computing platform with an emphasis on O-RAN (Open Radio Area Network) in particular, was announced by the Open Infrastructure Foundation today. Until until October 2020, the open source cloud computing initiative was known as the OpenStack Foundation. However, in an effort to extend its emphasis, the project changed its name to the Open Infrastructure Foundation. In order to allow edge computing for telecom use cases, the StarlingX project was founded in 2018 as a platform initiative that uses OpenStack as well as other open source cloud technologies like Kubernetes and Ceph.

Open Source NewsAnother day, another tweet—and who else but eccentric businessman Elon Musk? Elon suggested that Twitter's algorithm be made open-source in April 2022 to encourage "transparency in the platform." Even though he already owned Twitter when he made this claim, little has changed. In response to one of Elon's recent tweets calling for the platform to be open-source, an interested Twitter user said that it would be spectacular to accomplish. We can only wait at this point to watch how Elon and the Twitter team handle this particular situation and what occurs. Check out Elon's response above.