Skip to content

Latest commit

 

History

History
10 lines (8 loc) · 1.01 KB

README.md

File metadata and controls

10 lines (8 loc) · 1.01 KB

Modeling the Spatial Configuration of Everyday Objects using Natural Language

Final project for Math 111A, asking the question: can we use the similarity structure of our written language to model the spatial relationships of everyday objects in our environment?

preprocess_data.ipynb: reads in the .mat file from the NYU Depth Dataset V2, and processes the data into relevant .npy and .csv files. Creates the following files within data/:

  • depth_arr.npy: A (N, W, H) numpy array of depth clouds for each image in the sample.
  • image_arr.npy: A (N, W, H, 3) numpy array representing the RGB values of each image.
  • label_arr.npy: A (N, W, H) numpy array of object category labels, for each WxH pixel on the screen.
  • instance_arr.npy: A (N, W, H) numpy array of instance labels, for each WxH pixel on the screen.
  • metadata.csv: contains metadata for each image, including the image's scene ID, scene type, unique semantic labels, and unique instances.