Talk on pose estimation and audio-visual paper at ICPR 2020

With Ge Gao, we are giving a workshop talk at the ``Perception and Modelling for Manipulation of Objects’’ workshop organized at the 25th International Confererence on Pattern Recognition (ICPR 2020). Check out the talk if you are interested in applying deep learning on point cloud data for object pose estimation, generating synthetic training data for deep networks, and our ideas on how active and interactive perception can help in tasks such as object handover! The workshop takes place on January 10, 2020, and is held online.

Julius Richter from the Signal processing group at UHH is also presenting some joint work at ICPR, entitled ``Improving Mix-And-Separate Training in Audio-Visual Sound Source Separation with an Object Prior’’. You can find the paper online later, and the presentation details on the conference website, paper ID 1598.