This commit is contained in:
Stefano Flore
2024-10-17 10:38:45 +02:00
committed by GitHub
parent f7f6ca79b6
commit 436b0ebd95
2 changed files with 59 additions and 97 deletions
+58 -96
View File
@@ -15,131 +15,93 @@ Original project: https://github.com/THUDM/CogVideo
- Type: Multiline string.
- Impact: Directly influences the content and style of the generated video.
2. image (IMAGE)
### image (IMAGE)
- Description: The input image from which to start the video generation.
- Type: Image
- Impact: Serves as the starting point for the video, strongly influencing the first frames.
Type: Image
### num_frames (INT)
Impact: Serves as the starting point for the video, strongly influencing the first frames.
- Description: The total number of frames to generate.
- Default: 49
- Range: 49 - 2^31-1 (practically unlimited)
- Impact: Determines the length of the final video.
3. num_frames (INT)
### num_inference_steps (INT)
Description: The total number of frames to generate.
- Description: The number of inference steps for each frame.
- Default: 10
- Range: 1 - 1000
- Impact: Influences the quality and detail of each generated frame. More steps generally mean higher quality but longer processing times.
Default: 98
### guidance_scale (FLOAT)
Range: 49 - 2^31-1 (practically unlimited)
- Description: Controls how closely the model follows the prompt.
- Default: 6.0
- Range: 0.1 - 30.0
- Impact: Higher values produce results more faithful to the prompt but can lead to artifacts.
Step: 49
### use_dynamic_cfg (BOOLEAN)
Impact: Determines the length of the final video.
- Description: Enables or disables Dynamic Guided Configuration.
- Default: True
- Impact: When enabled, it can improve video consistency and quality.
4. num_inference_steps (INT)
### seed (INT)
Description: The number of inference steps for each frame.
- Description: The seed for random generation.
- Default: 0
- Range: 0 - 99999999999999
- Impact: Controls result reproducibility. The same seed will always produce the same output with the same parameters.
- Optical Flow Interpolation Parameters
Default: 10
### interpolation_factor (INT)
Range: 1 - 1000
- Description: Determines how many intermediate frames to create between each pair of original frames.
- Default: 1
- Range: 1 - 7 (steps of 2)
- Impact: Increases video smoothness by adding interpolated frames. A value of 1 doubles the number of frames, 3 quadruples it, etc.
Impact: Influences the quality and detail of each generated frame. More steps generally mean higher quality but longer processing times.
5. guidance_scale (FLOAT)
Description: Controls how closely the model follows the prompt.
Default: 6.0
Range: 0.1 - 30.0
Impact: Higher values produce results more faithful to the prompt but can lead to artifacts.
6. use_dynamic_cfg (BOOLEAN)
Description: Enables or disables Dynamic Guided Configuration.
Default: True
Impact: When enabled, it can improve video consistency and quality.
7. seed (INT)
Description: The seed for random generation.
Default: 0
Range: 0 - 99999999999999
Impact: Controls result reproducibility. The same seed will always produce the same output with the same parameters.
Optical Flow Interpolation Parameters
8. interpolation_factor (INT)
Description: Determines how many intermediate frames to create between each pair of original frames.
Default: 1
Range: 1 - 7 (steps of 2)
Impact: Increases video smoothness by adding interpolated frames. A value of 1 doubles the number of frames, 3 quadruples it, etc.
9. flow_precision (FLOAT)
### flow_precision (FLOAT)
- Description: Controls the level of detail in optical flow calculation.
- Default: 0.5
- Range: 0.1 - 1.0
- Impact: Higher values produce more precise optical flow but require more computation time.
Impact: Higher values produce more precise optical flow but require more computation time.
### motion_threshold (FLOAT)
10. motion_threshold (FLOAT)
- Description: Determines the minimum amount of movement required to apply interpolation.
- Default: 0.1
- Range: 0.0 - 1.0
- Impact: Lower values interpolate even small movements, higher values only significant movements.
Description: Determines the minimum amount of movement required to apply interpolation.
### smoothness (FLOAT)
Default: 0.1
- Description: Controls how "smooth" the interpolated movement should be.
- Default: 0.5
- Range: 0.0 - 1.0
- Impact: Higher values produce smoother transitions but may reduce movement details.
Range: 0.0 - 1.0
### flow_method (COMBO)
Impact: Lower values interpolate even small movements, higher values only significant movements.
- Description: The method used to calculate optical flow.
- Options: ["Farneback", "TV-L1", "DIS"]
- Impact: Each method has different characteristics in terms of accuracy and speed.
11. smoothness (FLOAT)
### edge_mode (COMBO)
Description: Controls how "smooth" the interpolated movement should be.
- Description: How to handle image edges during interpolation.
- Options: ["Replicate", "Reflect", "Wrap", "Constant"]
- Impact: Influences the appearance of edges in interpolated frames.
Default: 0.5
### interpolation_strength (FLOAT)
Range: 0.0 - 1.0
Impact: Higher values produce smoother transitions but may reduce movement details.
12. flow_method (COMBO)
Description: The method used to calculate optical flow.
Options: ["Farneback", "TV-L1", "DIS"]
Impact: Each method has different characteristics in terms of accuracy and speed.
13. edge_mode (COMBO)
Description: How to handle image edges during interpolation.
Options: ["Replicate", "Reflect", "Wrap", "Constant"]
Impact: Influences the appearance of edges in interpolated frames.
14. interpolation_strength (FLOAT)
Description: How heavily to apply the calculated interpolation.
Default: 1.0
Range: 0.0 - 1.0
Impact: Lower values produce subtler interpolation, higher values more pronounced interpolation.
- Description: How heavily to apply the calculated interpolation.
- Default: 1.0
- Range: 0.0 - 1.0
- Impact: Lower values produce subtler interpolation, higher values more pronounced interpolation.
## Updates
+1 -1
View File
@@ -177,7 +177,7 @@ class CogVideoXImageToVideoNodeExtended:
"required": {
"prompt": ("STRING", {"multiline": True}),
"image": ("IMAGE",),
"num_frames": ("INT", {"default": 98, "min": 49, "max": 2**31-1, "step": 49}),
"num_frames": ("INT", {"default": 49, "min": 49, "max": 2**31-1, "step": 49}),
"num_inference_steps": ("INT", {"default": 10, "min": 1, "max": 1000}),
"guidance_scale": ("FLOAT", {"default": 6.0, "min": 0.1, "max": 30.0}),
"use_dynamic_cfg": ("BOOLEAN", {"default": True}),