Big Data Set Design

Group: Yixuan Ye, Ruiyang Cheng, Yutong Dai

T-SNE is a nonlinear dimensionality reduction technique well-suited for high-dimensional data. For better display, we are going to visualize those results in a VR(Visual Reality) scene.
We use the images of LA river and relevant movies for example and designed the VR visualization experience for the tourists of Los Angeles. The database is from images of 5 locations in the LA river and 13 movies shot there. 
When people enter the VR scene, they can experience the machine learning process. By referring to the T-SNE results, he needs to judge if a random image from the movie image database is shot in a specific location among the 5 locations in the LA River.

Data: LA River

User: Film Lover Interested in LA River

Griffith Observatory

1st Street Bridge

4th Street Bridge

Long Beach

The National History Museum


Terminator II

The Dark Knight Rises

La La Land

Rebel Without a cause

LA River

©2019 by yixuanye. Proudly created with