Systems and methods for predicting future data using diverse sampling

A system for predicting future frames, comprising:

• a memory storing instructions that when executed by a processor cause the processor to:
• receive a video having a first frame from a first time and a second frame from a second time after the first time, wherein the first frame and the second frame are represented in an image space;
• update a frame prediction model based on the video;
• determine whether a stopping condition is satisfied;
• in response to determining that the stopping condition has been satisfied, generate a plurality of future frames for a third time after the second time, wherein the plurality of future frames is based on a normalized distance metric that preserves distance of samples in a latent space to the image space;
• and
• select a candidate frame from the plurality of future frames.

A non-transitory computer readable storage medium storing instructions that when executed by a computer having a processor to perform a method for predicting future frames, the method comprising:

• receiving a video having a first frame from a first time and a second frame from a second time after the first time, wherein the first frame and the second frame are represented in an image space;
• updating a frame prediction model based on the video;
• determining whether a stopping condition is satisfied;
• in response to determining that the stopping condition has been satisfied, generating a plurality of future frames for a third time after the second time, wherein the plurality of future frames is based on a normalized distance metric that preserves distance of samples in a latent space to the image space;
• and
• selecting a candidate frame from the plurality of future frames.