Jump to content

Assembly Line Activity Recognition: Difference between revisions

m
Line 311: Line 311:


In addition, the following plots show the training and validation loss for the best performing network, as well as the test confusion matrix; this matrix shows predominance along its diagonal; indicating a match between the networks predictions and the sample's ground truth.
In addition, the following plots show the training and validation loss for the best performing network, as well as the test confusion matrix; this matrix shows predominance along its diagonal; indicating a match between the networks predictions and the sample's ground truth.
 
<br>
<br>
<gallery widths=350px heights=350px mode=packed>
<gallery widths=350px heights=350px mode=packed>
File:Assembly-training-loss.svg|Loss plots
File:Assembly-training-loss.svg|Loss plots
Line 318: Line 319:


Finally, the inference demo below shows a gif of the assembly process under inference; it displays the top prediction for the sample as well as the associated softmax value. This inference process has a threshold of 0.5 meaning that if none of the softmax values achieve a 50% value, then no activity is recognized.
Finally, the inference demo below shows a gif of the assembly process under inference; it displays the top prediction for the sample as well as the associated softmax value. This inference process has a threshold of 0.5 meaning that if none of the softmax values achieve a 50% value, then no activity is recognized.
<br>
<br>
<br>
[[File:assembly-inference-results.gif|700px|center|thumb|Assembly inference results]]
[[File:assembly-inference-results.gif|700px|center|thumb|Assembly inference results]]
Cookies help us deliver our services. By using our services, you agree to our use of cookies.