https://github.com/mahnoorsheikh16/sketchify-a-quick-draw-drawing-classifier
https://github.com/mahnoorsheikh16/sketchify-a-quick-draw-drawing-classifier
bayesian-classifier drawing-classification feature-engineering gaussian-naive-bayes google-quick-draw kfold-cross-validation knn-classification linear-discriminant-analysis-lda logistic-regression matplotlib-pyplot multivariate-gaussian-distribution parzen-window pca-analysis rnn-pytorch seaborn-plots sequential-forward-selection spherical-gaussian-kernel svm-classifier tsne-visualization xgboost-classifier
Last synced: 15 days ago
JSON representation
- Host: GitHub
- URL: https://github.com/mahnoorsheikh16/sketchify-a-quick-draw-drawing-classifier
- Owner: mahnoorsheikh16
- Created: 2025-04-25T05:31:47.000Z (25 days ago)
- Default Branch: main
- Last Pushed: 2025-04-25T06:20:12.000Z (25 days ago)
- Last Synced: 2025-04-25T07:24:45.958Z (25 days ago)
- Topics: bayesian-classifier, drawing-classification, feature-engineering, gaussian-naive-bayes, google-quick-draw, kfold-cross-validation, knn-classification, linear-discriminant-analysis-lda, logistic-regression, matplotlib-pyplot, multivariate-gaussian-distribution, parzen-window, pca-analysis, rnn-pytorch, seaborn-plots, sequential-forward-selection, spherical-gaussian-kernel, svm-classifier, tsne-visualization, xgboost-classifier
- Homepage:
- Size: 290 KB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Sketchify – A Quick, Draw! drawing classifier
The project focuses on classifying digital drawings or doodles by identifying the object or category they represent.
*This is a course project for CSE 802 Pattern Recognition and Analysis at MSU. It is still ongoing so the code will be shared once the project is complete. (which will be very very soon. I don't like uploading incomplete things)
## Dataset:
The [Quick Draw Dataset](https://quickdraw.withgoogle.com/data) is a collection of 50 million drawings across 345 categories, contributed by players of the game [Quick, Draw!](https://quickdraw.withgoogle.com/). It has been compiled by Google Creative Lab to help developers train new neural networks, help researchers see patterns in how people around the world draw, and help artists create things we haven’t begun to think of.
The [simplified dataset](https://console.cloud.google.com/storage/browser/quickdraw_dataset/full/simplified;tab=objects?invt=AbvrVA&prefix=&forceOnObjectsSortingFiltering=false) consists of drawings recorded as timestamped vectors, tagged with metadata. It includes the following features:
1. `word`: category of the drawing
2. `countrycode`: player's country
3. `timestamp`
4. `recognized`: indicating whether the model successfully classified the drawing
5. `key_id`
6. `drawing array`: contains x and y pixel coordinates and t, which represents the time in milliseconds since the first point.