+- Selur's Little Message Board (https://forum.selur.net)
+-- Forum: Talk, Talk, Talk (https://forum.selur.net/forum-5.html)
+--- Forum: Small Talk (https://forum.selur.net/forum-7.html)
+--- Thread: Using Stable Diffision models for Colorization (/thread-4287.html)
RE: Using Stable Diffision models for Colorization - Selur - 12.06.2026
Just downloaded https://github.com/vapoursynth/vapoursynth/actions/runs/27424099595#artifacts ->
vapoursynth-win_amd64 which contains vapoursynth-77-cp312-abi3-win_amd64.whl
(running a test encode with BasicVSR++ atm.)
=> 5000 frame test encode ran without freezing.
Cu Selur
RE: Using Stable Diffision models for Colorization - Dan64 - 12.06.2026
I installed R77 over R76
And now I get the following error
2026-06-12 19:44:24.970
Failed to evaluate the script:
Python exception: No module named 'cv2'
Traceback (most recent call last):
File "vapoursynth.pyx", line 3623, in vapoursynth._vpy_evaluate
File "vapoursynth.pyx", line 3624, in vapoursynth._vpy_evaluate
File "D:\PProjects\vs-havc_dev\vs_test_debug_no-gpu\CMNET2DIT_test.vpy", line 15, in
import vshavc_test as havc
File "D:\PProjects\vs-havc_dev\vs_test_debug_no-gpu\vshavc_test\__init__.py", line 34, in
from vshavc_test.havc_utils import _get_tune_id, convert_format_RGB24, restore_format, HAVC_read_video, rgb_denoise
File "D:\PProjects\vs-havc_dev\vs_test_debug_no-gpu\vshavc_test\havc_utils.py", line 18, in
import cv2
ModuleNotFoundError: No module named 'cv2'Previously this script was working and cv2 is installed on my side (I just updated R76 with R77).
I will wait your release before confirm that the bug is fixed.
Dan
RE: Using Stable Diffision models for Colorization - Selur - 12.06.2026
'pip list' does list cv2 ? (at least I don't install cv/cv2 normally)
RE: Using Stable Diffision models for Colorization - Dan64 - 12.06.2026
cv2 is in the "opencv-python" package, and of course is installed (I also tried to reinstall)
Dan
RE: Using Stable Diffision models for Colorization - Selur - 12.06.2026
Strange.
RE: Using Stable Diffision models for Colorization - Dan64 - 14.06.2026
I tested vs_cmnet2dit() with the last dev version with R77 released by you today, and it works with no issues.
I think that the thread deadlock bug has been fixed.
Dan
RE: Using Stable Diffision models for Colorization - Dan64 - 14.06.2026
you can enable again (without core.num_threads=1) the filter CMNet2DiT which is currently disabled.
Dan
RE: Using Stable Diffision models for Colorization - Selur - 14.06.2026
Will do. (probably tomorrow)
RE: Using Stable Diffision models for Colorization - Selur - 15.06.2026
Uploaded a new dev version which enabled CMNet2DiT and removes the thread=1 limitation.
(uploading a new torch-add-on,.. upload will be done in ~ 1/2 hours)
Cu Selur
RE: Using Stable Diffision models for Colorization - Dan64 - 17.06.2026
As a natural extension of the "Fix Image" tab, I've added the new "Fix Video" tab.
Suppose you've already colorized a clip with CMNET2 and DitServerRPC. There will be a folder called "ref_qwen" containing the colored frames.
Suppose you've viewed the clip and want to change the color of some frames. Let's say these frames range from #34 to #525. After fixing them with the "Fix Image" tab, you can recolor the video only in the identified frame range (in this case, from #34 to #525). As shown in the image below.
I hope this helps; in some cases, it was necessary for me (which is why I added the implementation).
Since for technical reasons, I still need to re-encode the entire clip, it's mandatory to use as input the clip already colorized and to have NVEnC installed as explained in the README.
Dan