Computer Vision News - September 2021

3 Spotlight News 2 Designing Machine Learning Systems by Chip Huyen As our readers know, Chip Huyen is not only an awesome friend of our magazine, a very successful ML engineer and a popular ML lecturer at Stanford: she has also published a few traveling bestsellers in Vietnam. If you still don’t know her, read about her career in her own words. This is to tell you that US publisher O'Reilly is going to publish her English writing in this amazing guide: Designing Machine Learning Systems , a framework for designing real-world ML systems that are quick to deploy, reliable, scalable, and iterative. The first 3 chapters are already available here! GFP-GAN: Towards Real-World Blind Face Restoration with Generative Facial Prior Alessandro Ferrari kindly beckons we have a look at this: he says that Tencent AI just opened the source code of GFP-GAN, the most advanced photo restoration system ever! Blind face restoration usually relies on facial priors, such as facial geometry prior or reference prior, to restore realistic and faithful details. However, very low-quality inputs cannot offer accurate geometric prior while high-quality references are inaccessible, limiting the applicability in real-world scenarios. GFP-GAN could jointly restore facial details and enhance colors with just a single forward pass. Project Page AI Aids in Discovery of New Prognostic Biomarkers for Breast Cancer Scientists at Case Western Reserve University have used AI to identify new biomarkers for breast cancer that can predict whether the cancer will return after treatment - and which can be identified from routinely acquired tissue biopsy samples of early-stage breast cancer. We suppose this is not an exact prediction, however a very educated guess can be precious to determine correct diagnosis and treatment. Without entering into a discussion about correlation and causality, apparently the chances for an aggressive tumor and a likely recurrence are written in the arrangement of collagen in breast tissue. Read More