3D CNN 계산

3D CNN은 보통 Video 처리를 할 때 사용되며 output shape에 대한 계산은 다음과 같다.

The convolution formula is the same as in 2D and is well-described in CS231n tutorial:

where W is the input volume size, F is the receptive field size, S is the stride, and P is the amount of zero padding used on the border. In particular, when S=1 and P=0, like in your question, it simplifies to

So, if you input the tensor (40,64,64,12), ignoring the batch size, and F=3 then the output tensor size will be (38,62,62,8)

라고 한다.

3D CNN 계산에 대해 더 자세하게 정리된 글이 있어 참고의 블로그를 확인하자!!

참고:

• https://stats.stackexchange.com/questions/323313/how-to-calculate-output-shape-in-3d-convolution

• https://medium.com/@parkie0517/3d-convolution-%EC%99%84%EC%A0%84-%EC%A0%95%EB%B3%B5%ED%95%98%EA%B8%B0-using-pytorch-conv3d-4fab52c527d6 (3D CNN 계산 잘 돼있는 블로그)