Figure 3
From: Machine Learning-Based Analysis of Sperm Videos and Participant Data for Male Fertility Prediction
An illustration of how frames are stacked channel-wise after being greyscaled. From a video, a sequence of n frames are extracted and greyscaled. These frames are then stacked channel-wise, meaning each frame occupied one channel-dimension of the final image. The final stacked “image” is then of shape 224 × 224 × 30.
