Compute the Dice coefficient for the image prediction provided below. Note that black

box denotes the predicted bounding box, and red one denotes the ground truth

bounding box.

Round off your answer to the nearest ones place. For example, $57\mathrm{.}8\%$ will be $58\%.$

$76\%$

$84\%$

$91\%$

$96\%$U$-$Net

Which of the following elements distinguishes the U$-$Net framework from an encoder$-$decoder framework?

Convolutional layers

Skip connections

FC layers

Pooling layers