Skip to content

Commit 2037579

Browse files
authored
Update README.md
1 parent 44edda1 commit 2037579

File tree

1 file changed

+7
-0
lines changed

1 file changed

+7
-0
lines changed

README.md

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -1,13 +1,20 @@
11
# A-Method-Clustering-and-Classifying-Hand-shapes-from-Auslan
22

3+
## How to use
4+
5+
** Step 1 **
36
Use the get_signs.py script to create your data set or Auslan signs.
47

8+
** Step 2 **
59
Then use the create_data.py script to run the MediaPipe framework over your collected data to create a dataset for training a fully connected network.
610

11+
** Step 3 **
712
Next, use the make_hand_model.py script to train a full connected network. Be careful to examine the directory references.
813

14+
** Step 4 **
915
Finally, Use posenet_video.py script to classify your handshapes from data has been collected. Make sure this script references your trained hand model
1016

17+
## Optional
1118
For accuracy and performance metrics, use the classification_report.py script to see your models accuracy
1219

1320
Scripts that end in "clustering.py" and "crop_hand.py" are used for clustering from a video source, where the number of hand shapes are unknown.

0 commit comments

Comments
 (0)