Skip to content

9. Limitations & Future Scope

Rudra Barad edited this page Apr 1, 2021 · 1 revision
  • Proper Positioning of the face is required as the accuracy of face recognition also depends on the factors like brightness, facial expressions, etc.

  • We have allowed the program to run continuously without human intervention, so if a person stands in front of the camera for a long time (more than buffer time) then its counter will be incremented.

  • All the processes right now are serialized so that could be made parallelized so that a process doesn’t need to wait for all the processes to complete. And also, all the processes could be done at server side instead of the client-side to have centralized data for all the branches.

  • Accuracy is directly proportional to the available data set. There is always a scope to increase the accuracy of the models that we have used for predicting age, gender, and sentiments.

Clone this wiki locally