Recently I attended the International Conference on Computer Vision (ICCV) 2017 in Venice, Italy. Besides presenting a paper on Weakly Supervised Object Localization Using Things and Stuff Transfer, I was also one of the organizers of the Joint Workshop of the COCO and Places Challenges at ICCV 2017. I presented the COCO Stuff Segmentation Challenge in the main theatre of the Venice Film Festival (see image). The COCO Stuff Segmentation Challenge is a novel challenge that focuses on segmenting thing (car, person) and stuff (grass, sky) classes in an image. We annotated 55K (soon to be 160K!) images from the popular COCO dataset. After presenting the general outlines of the dataset and the challenge, I went on to present the top ranking submissions according to the public leaderboard. These were submission from Facebook, Google and the University of Oxford. The challenge winners then went on to give an in-depth overview over their methods. For further details see the presentation slides.