<div dir="ltr">Dear all,<div><br></div><div>Next week our CS seminar series will host Dr. Joao Carreira (Google DeepMind, London, <a href="http://home.isr.uc.pt/~joaoluis/">http://home.isr.uc.pt/~joaoluis/</a>) who will give a talk titled "Understanding people in videos". The talk will be held as usual in MB220 from 2pm to 3pm.</div><div><br></div><div>This will be the last talk before the Easter break, I hope you'll all be able to join! </div><div>Luca</div><div><br></div><div>Abstract:</div><div>The problem of “understanding” people in videos has been a long-standing central challenge in computer vision and artificial intelligence. In this talk I will first discuss recent technical advances in human pose estimation using models that make iterative passes through convolutional networks. I will then describe a novel (sorely needed) dataset for human action recognition, gathered from YouTube, having an order of magnitude more videos than existing datasets. I will show that this dataset enables a new type of spatiotemporal models which obtain results considerably above the state-of-the-art on popular benchmarks.</div><div><div><br></div>-- <br><div class="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div dir="ltr"><div dir="ltr">Luca Rossi</div><div dir="ltr"><br></div><div dir="ltr">Lecturer in Computer Science</div><div dir="ltr">School of Engineering and Applied Science</div><div dir="ltr">Aston University</div><div dir="ltr">Web: <a href="http://www.cs.bham.ac.uk/~rossil/" target="_blank">http://www.cs.aston.ac.uk/~rossil/</a></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
</div></div>