Hi!
I found that the "frame_concat" function in "once_devkit" can return the future multi-frame point cloud.
Does this mean that the past or future point cloud frames can also be used as the detection input at each moment?
Waymo enforced a causal system, for a frame at time step t, only sensor data up to time t can be used for its prediction.
Is it the same for ONCE?
Yes, you can use the past and future frames for detections on current frame if you want, there is no such constrains here.
Posted by: sslad2021 @ Sept. 26, 2021, 6:45 a.m.