Computer Vision News 4 A Great New Paper! Versatility is at the heart of POCO. “Our framework is agnostic to any other method that estimates human pose,” Sai explains. “For modeling uncertainty, we use conditional normalizing flow, a technique that takes a simpler distribution and converts it into a complex distribution.” POCO’s stellar team credits its success to a collaborative ethos, strategic partnerships fostered through initiatives like the European Laboratory for Learning and Intelligent Systems (ELLIS), and the support of organizations like the German Federal Ministry of Education and Research (BMBF). Along with forging strong ties with several ELLIS units around Europe, the team continues collaborating with INRIA Research Director Cordelia Schmid, a longtime MPI-IS friend. “Doing good research takes a village,” Dimitrios attests. “It’s very hard for isolated individuals to come
RkJQdWJsaXNoZXIy NTc3NzU=