Sneeze and cough detection with pytorch

Hi guys I am a newbie to pytorch so please bear with me if my question is wrong!
I googled lot about sneeze and cough detection but didnt get any useful resources. My main objective is to detect a person who sneezes or coughs with a live running camera with pytorch models along with torchvision or opencv.So can anyone please provide me with some examples , resources , source codes or links which I can refer to? Examples with Raspberry pi are most welcome as Iam planning to implement with it.

I hear that it is impossible to sneeze with eyes open. U can train a model to identify if a person has closed eyes or open eye. if his eyes are closed then u can look for further details to classify it as sneeze or something else

@Angry_potato is it possible in pytorch to train with audio files? Like classifying with the sound of sneezing or coughing?

Come on guys, do some basic research :smile:
Here are two papers using neural networks for cough detection: Cough detection using speech analysis and Deep Neural Networks for Identifying Cough Sounds. As I’m still a student, I’m able to access them through the VPN of my uni. I’m not sure if you can download the papers.

I misread the question, see bellow.

@boto there are lot of papers and theoretical approach based on this which I went through. I posted this question in the forum to get some help in the coding and implementation or hoping to get resources of datasets to train my model. So no one posts a question without good amount of research.

Oh damn, I misread your question. Sorry about that!
Unfortunately, I can’t help you with the implementation.

Hi! I saw your question on opencv forum as well. Have you had any progress so far? I was wondering if it could be implemented with thermal images taken from an IR camera.

Hi ,
Yes i have accomplished it using mel-spectograms. To differentiate different sounds