Hello, I want to ask the function of optical flow data.
In the literature, after input the pointcloud data into flownet3d, it can output the sceneflow. Therefore if we only need the pointcloud data to use the network, I don't understand why need to input optical flow data and gt data to get the sceneflow in training and evaluation? I think maybe it turns optical flow and gt into the correct sceneflow and makes the sceneflow as the label to train and test. Am I correct? If it is incorrect, what is the funtion of input optical flow and gt?
Hello, I want to ask the function of optical flow data. In the literature, after input the pointcloud data into flownet3d, it can output the sceneflow. Therefore if we only need the pointcloud data to use the network, I don't understand why need to input optical flow data and gt data to get the sceneflow in training and evaluation? I think maybe it turns optical flow and gt into the correct sceneflow and makes the sceneflow as the label to train and test. Am I correct? If it is incorrect, what is the funtion of input optical flow and gt?
Thank you.