https://github.com/adamspannbauer/ssbm_fox_detector

Keras object detector to detect Fox in Super Smash Bros Melee for the Nintendo Gamecube
https://github.com/adamspannbauer/ssbm_fox_detector

cnn cv frcnn keras object-detection python

Last synced: about 1 year ago
JSON representation

Keras object detector to detect Fox in Super Smash Bros Melee for the Nintendo Gamecube

• Host: GitHub
• URL: https://github.com/adamspannbauer/ssbm_fox_detector
• Owner: AdamSpannbauer
• License: mit
• Created: 2017-12-27T12:37:31.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2017-12-27T15:37:48.000Z (over 8 years ago)
• Last Synced: 2025-03-17T23:11:56.697Z (over 1 year ago)
• Topics: cnn, cv, frcnn, keras, object-detection, python
• Language: Python
• Homepage: https://adamspannbauer.github.io/2017/12/28/super-smash-cv/
• Size: 11.9 MB
• Stars: 37
• Watchers: 3
• Forks: 7
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# SSBM Fox Detector

## About

A [keras](https://keras.io/) FRCNN to detect Fox in Super Smash Bros Melee for the Nintendo Gamecube.  The FRCNN object detector was written by [yhenon](https://github.com/yhenon) and the orignal repo can be seen [here](https://github.com/yhenon/keras-frcnn).  The gameplay used to train and test the model is from VGBootcamp ([twitch](https://www.twitch.tv/vgbootcamp) & [youtube](https://www.youtube.com/channel/UCj1J3QuIftjOq9iv_rr7Egw)).  The model was trained on the [AMI provided by PyImageSearch](https://www.pyimagesearch.com/2017/09/20/pre-configured-amazon-aws-deep-learning-ami-with-python/).

[This google drive folder](https://drive.google.com/drive/folders/17QcuO9GQsiqO_4V86iV1gBGUCCfbPkUm) has the trained model file, `model_frcnn.hdf5`.  The folder also has the training data, `data.zip`, which contains a folder of images in addition to bounding box annotations produced by [dlib's imglab](https://github.com/davisking/dlib/tree/master/tools/imglab) (these annotations were converted to `fox_frcnn_tags.txt` for compatibility with the frcnn scripts).  Lastly the folder has a .zip containing images with bounding boxes drawn by the trained model, `results_imgs.zip`.

## Output

This gif shows the output of the model.  The model works with still images 1 at time; a sample of these single frame outputs were then converted into the gif.

![](readme/kjh_plup_vgbc_tbh7_fox_detect2.gif)

## Usage

If you'd like to use the trained fox model you'll need to:

*  clone this repo

*  download `model_frcnn.hdf5` to the cloned directory from [here](https://drive.google.com/drive/folders/17QcuO9GQsiqO_4V86iV1gBGUCCfbPkUm)

*  create a subdirectory named `results_imgs` in the cloned directory

*  run the command: `python test_frcnn.py -p /path/to/folder/with/images/to/annotate`

The model will output the annotated images to `results_imgs`.

*Note: the model was trained/tested only on blue (purple) melee fox*