let5see/AI-Video
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

模板中媒体尺寸改为预置方案

Browse Source
This commit is contained in:
puke
2025-11-12 17:19:06 +08:00
parent 64dcca204e
commit 7443cbf9c2
31 changed files with 576 additions and 90 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
api/routers/video.py
View File

@@ -73,8 +73,7 @@ async def generate_video_sync(
"max_narration_words": request_body.max_narration_words,
"min_image_prompt_words": request_body.min_image_prompt_words,
"max_image_prompt_words": request_body.max_image_prompt_words,
"image_width": request_body.image_width,
"image_height": request_body.image_height,
# Note: image_width and image_height are now auto-determined from template
"image_workflow": request_body.image_workflow,
"video_fps": request_body.video_fps,
"frame_template": request_body.frame_template,
@@ -161,8 +160,7 @@ async def generate_video_async(
"max_narration_words": request_body.max_narration_words,
"min_image_prompt_words": request_body.min_image_prompt_words,
"max_image_prompt_words": request_body.max_image_prompt_words,
"image_width": request_body.image_width,
"image_height": request_body.image_height,
# Note: image_width and image_height are now auto-determined from template
"image_workflow": request_body.image_workflow,
"video_fps": request_body.video_fps,
"frame_template": request_body.frame_template,

3
api/schemas/video.py
View File

@@ -57,8 +57,7 @@ class VideoGenerateRequest(BaseModel):
max_image_prompt_words: int = Field(60, ge=10, le=200, description="Max image prompt words")
# === Image Parameters ===
image_width: int = Field(1024, description="Image width")
image_height: int = Field(1024, description="Image height")
# Note: image_width and image_height are now auto-determined from template meta tags
image_workflow: Optional[str] = Field(None, description="Custom image workflow")
# === Video Parameters ===

7
pixelle_video/pipelines/custom.py
View File

@@ -92,8 +92,7 @@ class CustomPipeline(BasePipeline):
ref_audio: Optional[str] = None,
image_workflow: Optional[str] = None,
image_width: int = 1024,
image_height: int = 1024,
# Note: image_width and image_height are now auto-determined from template
frame_template: Optional[str] = None,
video_fps: int = 30,
@@ -161,6 +160,10 @@ class CustomPipeline(BasePipeline):
generator = HTMLFrameGenerator(template_path)
template_requires_image = generator.requires_image()
# Read media size from template meta tags
image_width, image_height = generator.get_media_size()
logger.info(f"📐 Media size from template: {image_width}x{image_height}")
if template_requires_image:
logger.info(f"📸 Template requires image generation")
else:

16
pixelle_video/pipelines/standard.py
View File

@@ -94,8 +94,7 @@ class StandardPipeline(BasePipeline):
max_image_prompt_words: int = 60,
# === Image Parameters ===
image_width: int = 1024,
image_height: int = 1024,
# Note: image_width and image_height are now auto-determined from template meta tags
image_workflow: Optional[str] = None,
# === Video Parameters ===
@@ -151,9 +150,8 @@ class StandardPipeline(BasePipeline):
min_image_prompt_words: Min image prompt length
max_image_prompt_words: Max image prompt length
image_width: Generated image width (default 1024)
image_height: Generated image height (default 1024)
image_workflow: Image workflow filename (e.g., "image_flux.json", None = use default)
Note: Image/video size is now auto-determined from template meta tags
video_fps: Video frame rate (default 30)
@@ -239,6 +237,16 @@ class StandardPipeline(BasePipeline):
template_config = self.core.config.get("template", {})
frame_template = template_config.get("default_template", "1080x1920/default.html")
# Read media size from template meta tags
from pixelle_video.services.frame_html import HTMLFrameGenerator
from pixelle_video.utils.template_util import resolve_template_path
template_path = resolve_template_path(frame_template)
temp_generator = HTMLFrameGenerator(template_path)
image_width, image_height = temp_generator.get_media_size()
logger.info(f"📐 Media size from template: {image_width}x{image_height}")
# Create storyboard config
config = StoryboardConfig(
task_id=task_id,

52
pixelle_video/services/frame_html.py
View File

@@ -141,6 +141,58 @@ class HTMLFrameGenerator:
logger.debug(f"Template loaded: {len(content)} chars")
return content
def _parse_media_size_from_meta(self) -> tuple[Optional[int], Optional[int]]:
"""
Parse media size from meta tags in template
Looks for meta tags:
- <meta name="template:media-width" content="1024">
- <meta name="template:media-height" content="1024">
Returns:
Tuple of (width, height) or (None, None) if not found
"""
from bs4 import BeautifulSoup
try:
soup = BeautifulSoup(self.template, 'html.parser')
# Find width and height meta tags
width_meta = soup.find('meta', attrs={'name': 'template:media-width'})
height_meta = soup.find('meta', attrs={'name': 'template:media-height'})
if width_meta and height_meta:
width = int(width_meta.get('content', 0))
height = int(height_meta.get('content', 0))
if width > 0 and height > 0:
logger.debug(f"Found media size in meta tags: {width}x{height}")
return width, height
return None, None
except Exception as e:
logger.warning(f"Failed to parse media size from meta tags: {e}")
return None, None
def get_media_size(self) -> tuple[int, int]:
"""
Get media size for image/video generation
Returns media size specified in template meta tags.
Returns:
Tuple of (width, height)
"""
media_width, media_height = self._parse_media_size_from_meta()
if media_width and media_height:
return media_width, media_height
# Fallback to default if not specified (should not happen with properly configured templates)
logger.warning(f"No media size meta tags found in template {self.template_path}, using fallback 1024x1024")
return 1024, 1024
def parse_template_parameters(self) -> Dict[str, Dict[str, Any]]:
"""
Parse custom parameters from HTML template

59
pixelle_video/services/video.py
View File

@@ -224,20 +224,43 @@ class VideoService:
-map "[v]" -map "[a]" output.mp4
"""
try:
inputs = [ffmpeg.input(v) for v in videos]
(
ffmpeg
.concat(*inputs, v=1, a=1)
.output(output)
.overwrite_output()
.run(capture_stdout=True, capture_stderr=True)
# Build filter_complex string manually
n = len(videos)
# Build input stream labels: [0:v][0:a][1:v][1:a]...
stream_spec = "".join([f"[{i}:v][{i}:a]" for i in range(n)])
filter_complex = f"{stream_spec}concat=n={n}:v=1:a=1[v][a]"
# Build ffmpeg command
cmd = ['ffmpeg']
for video in videos:
cmd.extend(['-i', video])
cmd.extend([
'-filter_complex', filter_complex,
'-map', '[v]',
'-map', '[a]',
'-y', # Overwrite output
output
])
# Run command
import subprocess
result = subprocess.run(
cmd,
capture_output=True,
text=True,
check=True
)
logger.success(f"Videos concatenated successfully: {output}")
return output
except ffmpeg.Error as e:
error_msg = e.stderr.decode() if e.stderr else str(e)
except subprocess.CalledProcessError as e:
error_msg = e.stderr if e.stderr else str(e)
logger.error(f"FFmpeg concat filter error: {error_msg}")
raise RuntimeError(f"Failed to concatenate videos: {error_msg}")
except Exception as e:
logger.error(f"Concatenation error: {e}")
raise RuntimeError(f"Failed to concatenate videos: {e}")
def _get_video_duration(self, video: str) -> float:
"""Get video duration in seconds"""
@@ -382,10 +405,17 @@ class VideoService:
# Concatenate original video with black padding
video_stream = ffmpeg.concat(video_stream, black_input.video, v=1, a=0)
# Prepare audio stream
# Prepare audio stream (pad if needed to match target duration)
input_audio = ffmpeg.input(audio)
audio_stream = input_audio.audio.filter('volume', audio_volume)
# Pad audio with silence if video is longer
if video_duration > audio_duration:
pad_duration = video_duration - audio_duration
logger.info(f"Video is longer, padding audio with {pad_duration:.2f}s silence")
# Use apad to add silence at the end
audio_stream = audio_stream.filter('apad', whole_dur=target_duration)
if not video_has_audio:
logger.info(f"Video has no audio stream, adding audio track")
# Video is silent, just add the audio
@@ -398,8 +428,7 @@ class VideoService:
output,
vcodec='libx264', # Re-encode video if padded
acodec='aac',
audio_bitrate='192k',
t=target_duration # Trim to target duration
audio_bitrate='192k'
)
.overwrite_output()
.run(capture_stdout=True, capture_stderr=True)
@@ -426,8 +455,7 @@ class VideoService:
output,
vcodec='libx264', # Re-encode video if padded
acodec='aac',
audio_bitrate='192k',
t=target_duration # Trim to target duration
audio_bitrate='192k'
)
.overwrite_output()
.run(capture_stdout=True, capture_stderr=True)
@@ -452,8 +480,7 @@ class VideoService:
output,
vcodec='libx264', # Re-encode video if padded
acodec='aac',
audio_bitrate='192k',
t=target_duration # Trim to target duration
audio_bitrate='192k'
)
.overwrite_output()
.run(capture_stdout=True, capture_stderr=True)

1
pyproject.toml
View File

@@ -25,6 +25,7 @@ dependencies = [
"uvicorn[standard]>=0.32.0",
"python-multipart>=0.0.12",
"comfykit>=0.1.9",
"beautifulsoup4>=4.14.2",
]
[project.optional-dependencies]

2
templates/1080x1080/minimal_framed.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1080, height=1080">
<title>极简边框风格 - 1080x1080</title>
<style>

2
templates/1080x1920/blur_card.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1080, height=1920">
<title>模糊背景卡片 - 1080x1920</title>
<!-- Google Fonts - 中文字体 -->

2
templates/1080x1920/cartoon.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>{{title}}</title>
<style>

2
templates/1080x1920/default.html
View File

@@ -2,6 +2,8 @@
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<style>
html {
margin: 0;

2
templates/1080x1920/elegant.html
View File

@@ -2,6 +2,8 @@
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<style>
html {
margin: 0;

2
templates/1080x1920/fashion_vintage.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1080, height=1920">
<title>时尚复古风格 - 1080x1920</title>
<style>

2
templates/1080x1920/full.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1080, height=1920">
<title>全屏图片 - 1080x1920</title>
<!-- Google Fonts - 中文字体 -->

2
templates/1080x1920/life_insights.html
View File

@@ -2,6 +2,8 @@
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<link rel="preconnect" href="https://fonts.googleapis.com">
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin>
<link href="https://fonts.googleapis.com/css2?family=Ma+Shan+Zheng&family=ZCOOL+KuaiLe&display=swap" rel="stylesheet">

2
templates/1080x1920/modern.html
View File

@@ -2,6 +2,8 @@
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<style>
html {
margin: 0;

2
templates/1080x1920/neon.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8" />
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=device-width, initial-scale=1" />
<title>{{title}}</title>
<style>

2
templates/1080x1920/psychology_card.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1080, height=1920">
<title>心理卡片风 - 1080x1920</title>
<style>

2
templates/1080x1920/purple.html
View File

@@ -2,6 +2,8 @@
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<style>
html {
margin: 0;

2
templates/1080x1920/simple.html
View File

@@ -2,6 +2,8 @@
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<style>
html {
margin: 0;

185
templates/1080x1920/video_simple.html Normal file
View File

@@ -0,0 +1,185 @@
<!DOCTYPE html>
<html>
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="512">
<meta name="template:media-height" content="288">
<style>
html {
margin: 0;
padding: 0;
height: 100%;
}
body {
margin: 0;
padding: 0;
width: 100%;
height: 100vh;
font-family: 'PingFang SC', 'Source Han Sans', 'Microsoft YaHei', sans-serif;
overflow: hidden;
/* background-color: #000; */
display: flex;
justify-content: center;
align-items: center;
}
/* 主容器 - 居中并包含所有内容 */
.main-container {
position: relative;
width: 1080px;
height: 1920px;
}
/* Background image layer (customizable using <img> tag) */
.background-image {
position: absolute;
top: 0;
left: 0;
width: 100%;
height: 100%;
z-index: 0;
}
/* Video overlay - 相对于main-container居中 */
.video-overlay {
position: absolute;
top: 50%;
left: 50%;
transform: translate(-50%, -50%);
width: 1080px;
height: 607px;
/* background: #f00; */
z-index: 1;
}
/* Title section - positioned above video */
.video-title-wrapper {
position: absolute;
top: calc(50% - 607px / 2 - 130px);
left: 50%;
transform: translateX(-50%);
max-width: 900px;
width: 900px;
text-align: center;
z-index: 2;
}
.video-title {
font-size: 72px;
font-weight: 700;
color: #ffffff;
line-height: 1.3;
letter-spacing: 3px;
text-shadow: 0 4px 20px rgba(0, 0, 0, 0.3);
margin-bottom: 20px;
}
/* 字幕区域 - 对齐视频底部 */
.content {
position: absolute;
bottom: calc(50% - 607px / 2 + 0px);
left: 50%;
transform: translateX(-50%);
width: 900px;
z-index: 4;
}
.text {
font-size: 40px;
color: #ffffff;
text-align: center;
line-height: 1.6;
font-weight: 500;
text-shadow:
2px 2px 4px rgba(0, 0, 0, 0.9),
0 0 8px rgba(0, 0, 0, 0.8),
0 0 16px rgba(0, 0, 0, 0.6);
padding: 10px 0px;
/* background-color: aqua; */
}
/* Footer - positioned below video */
.footer {
position: absolute;
top: calc(50% + 607px / 2 + 50px);
left: 50%;
transform: translateX(-50%);
width: 900px;
display: flex;
align-items: center;
justify-content: space-between;
padding-top: 40px;
border-top: 2px solid rgba(255, 255, 255, 0.3);
z-index: 2;
}
.author-section {
display: flex;
flex-direction: column;
gap: 8px;
}
.author {
font-size: 32px;
font-weight: 600;
color: #ffffff;
text-shadow: 0 2px 8px rgba(0, 0, 0, 0.2);
}
.author-desc {
font-size: 24px;
color: rgba(255, 255, 255, 0.9);
font-weight: 400;
}
.logo-section {
display: flex;
flex-direction: column;
align-items: flex-end;
gap: 10px;
}
.logo {
font-size: 28px;
font-weight: 600;
color: #ffffff;
letter-spacing: 2px;
text-shadow: 0 2px 8px rgba(0, 0, 0, 0.2);
}
</style>
</head>
<body>
<!-- 主容器 - 所有元素都在这里面，相对于video-overlay定位 -->
<div class="main-container">
<!-- Background image layer (customizable via background parameter) -->
<div class="background-image">
</div>
<!-- Video overlay - 居中参考点 -->
<div class="video-overlay"></div>
<!-- Video title - positioned above video -->
<div class="video-title-wrapper">
<div class="video-title">{{title}}</div>
</div>
<!-- 字幕区域 - 独立定位在视频底部 -->
<div class="content">
<div class="text">{{text}}</div>
</div>
<!-- Footer - positioned below video -->
<div class="footer">
<div class="author-section">
<div class="author">{{author=@Pixelle.AI}}</div>
<div class="author-desc">{{describe=Open Source Omnimodal AI Creative Agent}}</div>
</div>
<div class="logo-section">
<div class="logo">{{brand=Pixelle-Video}}</div>
</div>
</div>
</div>
</body>
</html>

2
templates/1920x1080/film.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1920, height=1080">
<title>视频模板 - 电影风格</title>
<style>

2
templates/1920x1080/full.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1920, height=1080">
<title>全屏图片 - 1920x1080</title>
<!-- Google Fonts - 中文字体 -->

2
templates/1920x1080/ultrawide_minimal.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1920, height=1080">
<title>视频模板 - 极简风格</title>
<style>

2
templates/1920x1080/wide_darktech.html
View File

@@ -2,6 +2,8 @@
<html lang="zh-CN">
<head>
<meta charset="UTF-8">
<meta name="template:media-width" content="1024">
<meta name="template:media-height" content="1024">
<meta name="viewport" content="width=1920, height=1080">
<title>视频模板 - 横屏科技风格</title>
<style>

24
uv.lock generated
View File

@@ -226,6 +226,19 @@ wheels = [
{ url = "https://files.pythonhosted.org/packages/2f/eb/f25ad1a7726b2fe21005c3580b35fa7bfe09646faf7c8f41867747987a35/beartype-0.22.4-py3-none-any.whl", hash = "sha256:7967a1cee01fee42e47da69c58c92da10ba5bcfb8072686e48487be5201e3d10", size = 1318387 },
]
[[package]]
name = "beautifulsoup4"
version = "4.14.2"
source = { registry = "https://pypi.org/simple" }
dependencies = [
{ name = "soupsieve" },
{ name = "typing-extensions" },
]
sdist = { url = "https://files.pythonhosted.org/packages/77/e9/df2358efd7659577435e2177bfa69cba6c33216681af51a707193dec162a/beautifulsoup4-4.14.2.tar.gz", hash = "sha256:2a98ab9f944a11acee9cc848508ec28d9228abfd522ef0fad6a02a72e0ded69e", size = 625822 }
wheels = [
{ url = "https://files.pythonhosted.org/packages/94/fe/3aed5d0be4d404d12d36ab97e2f1791424d9ca39c2f754a6285d59a3b01d/beautifulsoup4-4.14.2-py3-none-any.whl", hash = "sha256:5ef6fa3a8cbece8488d66985560f97ed091e22bbc4e9c2338508a9d5de6d4515", size = 106392 },
]
[[package]]
name = "blinker"
version = "1.9.0"
@@ -1653,6 +1666,7 @@ name = "pixelle-video"
version = "0.1.2"
source = { editable = "." }
dependencies = [
{ name = "beautifulsoup4" },
{ name = "certifi" },
{ name = "comfykit" },
{ name = "edge-tts" },
@@ -1680,6 +1694,7 @@ dev = [
[package.metadata]
requires-dist = [
{ name = "beautifulsoup4", specifier = ">=4.14.2" },
{ name = "certifi", specifier = ">=2025.10.5" },
{ name = "comfykit", specifier = ">=0.1.9" },
{ name = "edge-tts", specifier = ">=7.2.3" },
@@ -2461,6 +2476,15 @@ wheels = [
{ url = "https://files.pythonhosted.org/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2", size = 10235 },
]
[[package]]
name = "soupsieve"
version = "2.8"
source = { registry = "https://pypi.org/simple" }
sdist = { url = "https://files.pythonhosted.org/packages/6d/e6/21ccce3262dd4889aa3332e5a119a3491a95e8f60939870a3a035aabac0d/soupsieve-2.8.tar.gz", hash = "sha256:e2dd4a40a628cb5f28f6d4b0db8800b8f581b65bb380b97de22ba5ca8d72572f", size = 103472 }
wheels = [
{ url = "https://files.pythonhosted.org/packages/14/a0/bb38d3b76b8cae341dad93a2dd83ab7462e6dbcdd84d43f54ee60a8dc167/soupsieve-2.8-py3-none-any.whl", hash = "sha256:0cc76456a30e20f5d7f2e14a98a4ae2ee4e5abdc7c5ea0aafe795f344bc7984c", size = 36679 },
]
[[package]]
name = "sse-starlette"
version = "3.0.3"

61
web/app.py
View File

@@ -782,6 +782,11 @@ def main():
generator_for_params = HTMLFrameGenerator(template_path_for_params)
custom_params_for_video = generator_for_params.parse_template_parameters()
# Get media size from template (for image/video generation)
media_width, media_height = generator_for_params.get_media_size()
st.session_state['template_media_width'] = media_width
st.session_state['template_media_height'] = media_height
# Detect template media type
from pathlib import Path
template_name = Path(frame_template).name
@@ -1023,43 +1028,18 @@ def main():
else:
workflow_key = "runninghub/image_flux.json" # fallback
# Get media size from template
image_width = st.session_state.get('template_media_width', 1024)
image_height = st.session_state.get('template_media_height', 1024)
# Display media size info (read-only)
if template_media_type == "video":
size_info_text = tr('style.video_size_info', width=image_width, height=image_height)
else:
size_info_text = tr('style.image_size_info', width=image_width, height=image_height)
st.info(f"📐 {size_info_text}")
# 2. Media size input
col1, col2 = st.columns(2)
with col1:
if template_media_type == "video":
width_label = tr('style.video_width')
width_help = tr('style.video_width_help')
else:
width_label = tr('style.image_width')
width_help = tr('style.image_width_help')
image_width = st.number_input(
width_label,
min_value=128,
value=1024,
step=1,
label_visibility="visible",
help=width_help
)
with col2:
if template_media_type == "video":
height_label = tr('style.video_height')
height_help = tr('style.video_height_help')
else:
height_label = tr('style.image_height')
height_help = tr('style.image_height_help')
image_height = st.number_input(
height_label,
min_value=128,
value=1024,
step=1,
label_visibility="visible",
help=height_help
)
# 3. Prompt prefix input
# Prompt prefix input
# Get current prompt_prefix from config
current_prefix = comfyui_config["image"]["prompt_prefix"]
@@ -1152,10 +1132,12 @@ def main():
st.info("ℹ️ " + tr("image.not_required"))
st.caption(tr("image.not_required_hint"))
# Get media size from template (even though not used, for consistency)
image_width = st.session_state.get('template_media_width', 1024)
image_height = st.session_state.get('template_media_height', 1024)
# Set default values for later use
workflow_key = None
image_width = 1024
image_height = 1024
prompt_prefix = ""
@@ -1225,14 +1207,13 @@ def main():
progress_bar.progress(min(int(event.progress * 100), 99)) # Cap at 99% until complete
# Generate video (directly pass parameters)
# Note: image_width and image_height are now auto-determined from template
video_params = {
"text": text,
"mode": mode,
"title": title if title else None,
"n_scenes": n_scenes,
"image_workflow": workflow_key,
"image_width": int(image_width),
"image_height": int(image_height),
"frame_template": frame_template,
"prompt_prefix": prompt_prefix,
"bgm_path": bgm_path,

13
web/i18n/locales/en_US.json
View File

@@ -49,17 +49,8 @@
"style.workflow_how": "Place the exported image_xxx.json workflow file(API format) into the workflows/selfhost/ folder (for local ComfyUI) or the workflows/runninghub/ folder (for cloud)",
"style.video_workflow_what": "Determines how each frame's video clip is generated and its effect (e.g., using different video generation models)",
"style.video_workflow_how": "Place the exported video_xxx.json workflow file(API format) into the workflows/selfhost/ folder (for local ComfyUI) or the workflows/runninghub/ folder (for cloud)",
"style.image_size": "Image Size",
"style.image_width": "Width",
"style.image_height": "Height",
"style.image_width_help": "Width of AI-generated images (Note: This is the image size, not the final video size. Video size is determined by the template)",
"style.image_height_help": "Height of AI-generated images (Note: This is the image size, not the final video size. Video size is determined by the template)",
"style.video_width": "Video Width",
"style.video_height": "Video Height",
"style.video_width_help": "Width of AI-generated video (Note: This is the video clip size, will auto-adapt to template size)",
"style.video_height_help": "Height of AI-generated video (Note: This is the video clip size, will auto-adapt to template size)",
"style.image_size_note": "Image size controls the dimensions of AI-generated illustrations, and does not affect the final video size. Video size is determined by the Storyboard Template below.",
"style.video_size_note": "Video size will automatically adapt to the template size, no manual adjustment needed.",
"style.image_size_info": "Image Size: {width}x{height} (auto-determined by template)",
"style.video_size_info": "Video Size: {width}x{height} (auto-determined by template)",
"style.prompt_prefix": "Prompt Prefix",
"style.prompt_prefix_what": "Automatically added before all image prompts to control the illustration style uniformly (e.g., cartoon, realistic)",
"style.prompt_prefix_how": "Enter style description in the input box below. To save permanently, edit the config.yaml file",

13
web/i18n/locales/zh_CN.json
View File

@@ -49,17 +49,8 @@
"style.workflow_how": "将导出的 image_xxx.json 工作流文件（API格式）放入 workflows/selfhost/（本地 ComfyUI）或 workflows/runninghub/（云端）文件夹",
"style.video_workflow_what": "决定视频中每帧视频片段的生成方式和效果（如使用不同的视频生成模型）",
"style.video_workflow_how": "将导出的 video_xxx.json 工作流文件（API格式）放入 workflows/selfhost/（本地 ComfyUI）或 workflows/runninghub/（云端）文件夹",
"style.image_size": "图片尺寸",
"style.image_width": "宽度",
"style.image_height": "高度",
"style.image_width_help": "AI 生成插图的宽度（注意：这是插图尺寸，不是最终视频尺寸。视频尺寸由模板决定）",
"style.image_height_help": "AI 生成插图的高度（注意：这是插图尺寸，不是最终视频尺寸。视频尺寸由模板决定）",
"style.video_width": "视频宽度",
"style.video_height": "视频高度",
"style.video_width_help": "AI 生成视频的宽度（注意：这是视频片段尺寸，会自适应模板尺寸）",
"style.video_height_help": "AI 生成视频的高度（注意：这是视频片段尺寸，会自适应模板尺寸）",
"style.image_size_note": "图片尺寸控制 AI 生成的插图大小，不影响最终视频尺寸。视频尺寸由下方的「📐 分镜模板」决定。",
"style.video_size_note": "视频尺寸会自动适配模板尺寸，无需手动调整。",
"style.image_size_info": "插图尺寸：{width}x{height}（由模板自动决定）",
"style.video_size_info": "视频尺寸：{width}x{height}（由模板自动决定）",
"style.prompt_prefix": "提示词前缀",
"style.prompt_prefix_what": "自动添加到所有图片提示词前面，统一控制插图风格（如：卡通风格、写实风格等）",
"style.prompt_prefix_how": "直接在下方输入框填写风格描述。若要永久保存，需编辑 config.yaml 文件",

5
workflows/runninghub/video_wan2.1_fusionx.json Normal file
View File

@@ -0,0 +1,5 @@
{
"source": "runninghub",
"workflow_id": "1985909483975188481"
}

187
workflows/selfhost/video_wan2.1_fusionx.json Normal file
View File

@@ -0,0 +1,187 @@
{
"3": {
"inputs": {
"seed": 576600626757621,
"steps": 10,
"cfg": 1,
"sampler_name": "uni_pc",
"scheduler": "normal",
"denoise": 1,
"model": [
"48",
0
],
"positive": [
"6",
0
],
"negative": [
"7",
0
],
"latent_image": [
"40",
0
]
},
"class_type": "KSampler",
"_meta": {
"title": "KSampler"
}
},
"6": {
"inputs": {
"text": [
"49",
0
],
"clip": [
"38",
0
]
},
"class_type": "CLIPTextEncode",
"_meta": {
"title": "CLIP Text Encode (Positive Prompt)"
}
},
"7": {
"inputs": {
"text": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
"clip": [
"38",
0
]
},
"class_type": "CLIPTextEncode",
"_meta": {
"title": "CLIP Text Encode (Negative Prompt)"
}
},
"8": {
"inputs": {
"samples": [
"3",
0
],
"vae": [
"39",
0
]
},
"class_type": "VAEDecode",
"_meta": {
"title": "VAE Decode"
}
},
"30": {
"inputs": {
"frame_rate": 16,
"loop_count": 0,
"filename_prefix": "Video",
"format": "video/h264-mp4",
"pix_fmt": "yuv420p",
"crf": 19,
"save_metadata": true,
"trim_to_audio": false,
"pingpong": false,
"save_output": true,
"images": [
"8",
0
]
},
"class_type": "VHS_VideoCombine",
"_meta": {
"title": "Video Combine 🎥🅥🅗🅢"
}
},
"37": {
"inputs": {
"unet_name": "wan-fusionx/WanT2V_MasterModel.safetensors",
"weight_dtype": "default"
},
"class_type": "UNETLoader",
"_meta": {
"title": "Load Diffusion Model"
}
},
"38": {
"inputs": {
"clip_name": "umt5_xxl_fp8_e4m3fn_scaled.safetensors",
"type": "wan",
"device": "default"
},
"class_type": "CLIPLoader",
"_meta": {
"title": "Load CLIP"
}
},
"39": {
"inputs": {
"vae_name": "wan_2.1_vae.safetensors"
},
"class_type": "VAELoader",
"_meta": {
"title": "Load VAE"
}
},
"40": {
"inputs": {
"width": [
"50",
0
],
"height": [
"51",
0
],
"length": 81,
"batch_size": 1
},
"class_type": "EmptyHunyuanLatentVideo",
"_meta": {
"title": "EmptyHunyuanLatentVideo"
}
},
"48": {
"inputs": {
"shift": 1,
"model": [
"37",
0
]
},
"class_type": "ModelSamplingSD3",
"_meta": {
"title": "Shift"
}
},
"49": {
"inputs": {
"value": "草地上有个小狗在奔跑"
},
"class_type": "PrimitiveStringMultiline",
"_meta": {
"title": "$prompt.value!"
}
},
"50": {
"inputs": {
"value": 512
},
"class_type": "easy int",
"_meta": {
"title": "$width.value"
}
},
"51": {
"inputs": {
"value": 288
},
"class_type": "easy int",
"_meta": {
"title": "$height.value"
}
}
}