Can't get Text2Vid working with a reference video

[kaushal101]

Hi All,
I must be missing something simple here. I'm able to make the Vid2Vid work fine, with controlnet, without any issue.

I want the basic movement of the subject same in output, but the look and feel quite different. So I am trying to use text2vid, with a reference video.

When I do this, the first frame of the video gets generated fine, with controlnet working.

But after the first frame, controlnet doesn't receive any reference frames. Ideally each frame from reference video should be supplied to controlnet. But that stops working after the first frame.

From frame 2, I start receiving below error.
2023-10-03 17:24:46,744 - ControlNet - WARNING - No ControlNetUnit detected in args. It is very likely that you are having an extension conflict.Here are args received by ControlNet: [False, None, None].
2023-10-03 17:24:46,757 - ControlNet - WARNING - No ControlNetUnit detected in args. It is very likely that you are having an extension conflict.Here are args received by ControlNet: [False, None, None].

I have correctly set the parameter "What frame should be sent to CN" (parameter cn_frame_send) to 3 (Current reference video frame), but it just refuses to work.

Am I missing something simple here? Any help would be greatly appreciated.

Thanks in advance!