We show our prediction results and baseline prediction results on 15 tasks for PHYRE (within-task generalization, fold 0).
Given the first image and the objects' bounding boxes, we predict the future bounding boxes and masks for each moving objects. The backgrounds (colored purple and black) is taken from the input image.
We show our prediction results on PHYRE (cross-task generalization).
Given the first image and the objects' bounding boxes, we predict the future bounding boxes and masks for each moving objects. The backgrounds (colored purple and black) is taken from the input image. The image shown here is not seen during training.
We show our predicted and ground-truth trajectory below. We visualize the center location of our predicted bounding boxes.
We show our predicted and ground-truth trajectory below. We visualize the center location of our predicted bounding boxes.
We show our predicted and ground-truth trajectory below. We visualize the center location of our predicted bounding boxes.