Data preprep:
rename_data -> resize_data -> strip_data
Network: nate_regression_redux with data_loader
Appendix:
dataset_full: folder with data
Set01: set of 320 images of 32 colors
exif01: metadata from images of set01, only the exposure time and iso is consistant
lab_colors01: colors for set01 from colorreader, in Lab, the first row is the corresponding number of images per color
Set02: set of 24 images of 12 different colors
exif02: metadata from images of set02, only the exposure time and iso is consistant
lab_colors02: colors for set02 from colorreader, in Lab, the first row is the corresponding number of images per color
project_documents: all the documents from the project
Group10_Final_Project.pdf: final presentation
Vishion Literature Review.pdf: literature review of related research
Vishion Market Research.pdf: exploration of our customer and their market
Vishion Technical Plan.pdf: how we planned this repository
slideshow_images: images from the slideshow
README.md: You're reading it!
classification_data_loader.py: for nates_classification, imports the data a bit differently with 32 classfications instead of color values
color_extraction.py: k means algorithm for extracting dominant color and color name detector
connected_component.py: grab cut mask shape changed
data_loader.py: loads data for nates_regression and nates_regression_redux, imports the data with raw colors
data_visualization.py: creates chart of color classifications (used in presentation)
nates_classification.py: classifcation algorithm
nates_regression.py: simple regression network
nates_regression_redux.py: more complex regression network
rename_data.py: dataset prep (renames files)
resize_data.py: dataset prep (resized image)
shweta_kmeans_color_extractor.py: k means clustering code changes 17 hours ago
simple_color_detector.py: k means algorithm for extracting dominant color and color name detector a day ago
strip_data.py: dataset prep (copies metadata to exif0*)