Intrigued by how reflection changes images in subtle and not-so-subtle ways, a team of researchers used artificial intelligence to investigate what sets originals apart from their reflections. Their algorithms learned to pick up on unexpected clues such as hair parts, gaze direction and, surprisingly, beards - findings with implications for training machine learning models and detecting faked images.
source https://www.sciencedaily.com/releases/2020/07/200702152445.htm
Friday, 3 July 2020
Related Posts
More filling? Tastes great? How flies, and maybe people, choose their foodFlies have discriminating taste. Like a gourmet perusing a menu, they … Read More
High-throughput metabolic profiling of single cellsScientists have presented a new method for generating metabolic profil… Read More
Machine learning cracks the oxidation states of crystal structuresChemical engineers have developed a machine-learning model that can pr… Read More
New study presents transformative metasurface based on zerogap embedded templateScientists have developed a new technique of predefining the crack pat… Read More
Bitumen surfaces under the microscopeAlthough the history of bitumen dates back to the third millennium BC,… Read More
New species of pseudo-horses living 37 million years agoScientists have described two new species of palaeotheriidae mammals t… Read More
0 comments: