Figure 2
The collage contains three labeled columns, each with three pictures and descriptive text above them. The first column is labeled “A very large pizza covered in cheese and toppings”, showing three pizzas on plates or trays. The second column is labeled “A giraffe in a field with trees in the background”, showing giraffes standing in open fields with trees behind them. The third column is labeled “A skier is in the snow going downhill”, showing skiers dressed in winter clothing gliding down snowy slopes.Images generated by the same model architecture, trained on the MS-COCO dataset [9]