Blur in a specific area with Opencv

Question

I am trying to blur a face found in a video using Opencv but am having the following error:

cv2.error: Opencv(4.1.0) /io/opencv/modules/imgproc/src/filter.simd.hpp:220: error: (-215:Assertion failed) src && dst && Count > 0 in Function 'Filterengine__proceed

I ran another code with the same logic changing the input to a static image and works perfectly.

Line in which the error occurs:

frame[x:x+w, y:y+h] = cv2.Blur(frame[x:x+w, y:y+h], (23, 23))

Complete code:

import cv2
import sys
import numpy

#argumento para entrada de arquivos de video
video = sys.argv[1]
#argumento para entrada de classificadores
cascPath = sys.argv[2]

#carrega o arquivo de classificacao treinado
faceCascades = cv2.CascadeClassifier(cascPath)

#capturando o video
cap = cv2.VideoCapture(video)

while(1):
    #lendo a imagem
    ret, frame = cap.read()
    #passando a imagem para cinza
    gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)

    #localizando faces
    faces = faceCascades.detectMultiScale(
        gray,
        scaleFactor=1.1,
        minNeighbors=5,
        minSize=(30,30),
        flags=cv2.CASCADE_SCALE_IMAGE
    )

    for (x, y, w, h) in faces:
        frame[x:x+w, y:y+h] = cv2.blur(frame[x:x+w, y:y+h], (23, 23))

        #exibindo o frame processado
        cv2.imshow('frame', frame)


    #aguardando tecla para finalizar o programa
    if cv2.waitKey(1) == ord('q'):
        break

#destroi todas janelas abertas 
cv2.destroyAllWindows()

I couldn’t find any more detailed content on Opencv besides the documentation, so if anyone has a good reference of studies on the subject, it will be very welcome.