diff --git a/index.html b/index.html index f3dd848..400d312 100644 --- a/index.html +++ b/index.html @@ -122,26 +122,29 @@
MarioVGG is a text-to-video model which takes in the initial game state as a single image frame, and an action in the form of a text prompt. It then outputs a video (a sequence of frames) depicting the action in the game.
+Comparison between MarioVGG-generated frames and ground truth frames.
+To generate and maintain gameplay for longer horizons, we chain together multiple model generations recursively.