This line should be np.float32.

[peifferjd]

https://github.com/peabody124/PosePipeline/blob/c987d4348c4f884f5534d95121d27e9a3ebb68ed/pose_pipeline/utils/bounding_box.py#L50

np.float32 is still able to be used here and actually needs to be. The standard float gives an error.