CogVideo:重磅升级!图生视频完美镜头控制和3D环绕,商用级开源AI视频曙光

科技   2024-11-12 21:47   浙江  

CogVideo:重磅升级!图生视频完美镜头控制和3D环绕,商用级开源AI视频曙光

🌹大家好!欢迎来到破狼公众号。感谢大家的支持与鼓励。在AIGC探索道路上,我将与你一路同行。喜欢就星标关注破狼公众号或文末扫码加入交流群 !

CogVideo简介

在之前的文章已经多次介绍过智普开源的视频生成模型:CogVideoX。在当前AI视频开源领域主要为MochiCogVideoX两大主力。在社区生态的快速迭代中,两大视频模型视频生成质量也是快速的提升。Mochi从开源初的2张H100 显存要求在社区量化等优化手段下,如果已经可以在4090消费级显卡上推理运行了。而CogVideoX也发布了最新的1.5版本支持不仅支持文生视频和图生视频,当前图生视频也支持了任意长宽比的视频生成了。以及CogVideoX还开放了LORA体系,社区也逐步的推出了首批CogVideoX LORA模型。今天文章的主题就聚焦中CogVideoX首批LORA模型中代表模型:运镜模型。当视频模型涌现大量LORA和ControlNet那距离成熟已不远(期待开源视频界Flux-Dev)。

  • • CogVideo:https://github.com/THUDM/CogVideo

  • • DimensionX运镜

    https://huggingface.co/wenqsun/DimensionX/tree/main

更多参考资料

CogVideo运镜 ComfyUI体验

首先需要更新ComfyUI并通过插件管理器Git安装ComfyUI-CogVideoXWrapper,模型会在首次运行时候自动下载。注意选择图生视频模型:CogVideoX-5b-I2V

  • • 插件地址:https://github.com/kijai/ComfyUI-CogVideoXWrapper

  • • CogVideoX-5b-I2V模型:模型会在首次自动下载,如需手动下载模型,则需要下载整个项目全部文件,并放置/ComfyUI/models/CogVideo/CogVideoX-5b-I2V。下载地址:https://huggingface.co/THUDM/CogVideoX-5b-I2V

  • • orbit_left_lora_weights.safetensors和orbit_up_lora_weights.safetensors:系在运镜LORA模型并放置到目录/ComfyUI/models/CogVideo/loras下。下载地址:文末网盘获取,或

    https://huggingface.co/wenqsun/DimensionX/tree/main

Flux文生图工作流

Flux文生图感兴趣的同学可参考LIBLIB在线运行工作流:FLUX[续篇]:12B参数23G最大开源文生图模型,Dev版直出惊艳美图欣赏

本文涉及ComfyUI工作流和模型均可在LIBLIBAI上下载或在线运行体验:
• FLUX.1哩布在线可运行-黑暗森林工作室
https://www.liblib.art/modelinfo/488cd9d58cd4421b9e8000373d7da123
• F.1-绮梦流光-水湄凝香
https://www.liblib.art/modelinfo/134c6dd95aef48e98a22b24e003e026b
• 工作流-Flux文|图生图+LORA+提示反推一键切换工作流
https://www.liblib.art/modelinfo/782aacd70f604da39e83368c696a02a8

另外LIBLIBAI已支持本地客户端使用可首页下载体验。

CogVideo运镜工作流

CogVideo运镜工作流已上传LIBLIB平台

https://www.liblib.art/modelinfo/a1591ac2fad94bd38739af73966f6ce6?versionUuid=2fc19a091b7e4374987b0a76aa033627

注意

  • • 这里新增了一个CogVideoLoraSelectLORA节点,作为运镜LORA加载器
  • • CogVideoLoraSelect节点的fuse_lora需要设置为trueCogVideoDecode节点的enable_vae_tiling设置为true减少显存使用。另外如果低显存可启用fp8_transformer

01. 猛犸象群

Several giant wooly mammoths approach treading through a snowy meadow, their long wooly fur lightly blows in the wind as they walk, snow covered trees and dramatic snow capped mountains in the distance, mid afternoon light with wispy clouds and a sun high in the distance creates a warm glow, the low camera view is stunning capturing the large furry mammal with beautiful photography, depth of field.

02.城市上空游动的鲸鱼

Whales swim in the air. In the mesmerizing nightscape, a colossal whale glides gracefully through the star-studded sky, its vast, textured body illuminated by the soft, ethereal glow of the moon. The city below, a sprawling metropolis of towering skyscrapers, twinkles with countless lights, creating a captivating contrast between the urban jungle and the serene marine giant. The sky, painted in deep shades of blue and adorned with twinkling stars, adds a dreamlike quality to the scene. The whale, seemingly in motion, appears to be swimming through the clouds, its majestic form a surreal and awe-inspiring sight against the backdrop of the illuminated cityscape

03.鸡尾酒调制

In a cozy, well-lit kitchen, a man in a black apron and blue cap is meticulously crafting a cocktail. He stands behind a white countertop, expertly pouring a rich, amber liquid from a shaker into a martini glass. The scene is filled with various bottles of alcohol, a juicer, and other bar tools, indicating a well-equipped home bar. The window behind him reveals a serene suburban view, adding a touch of calm to the focused atmosphere. His precise movements and the array of ingredients suggest a passion for mixology, creating a moment of artistry in an everyday setting.

04.3D镜头旋转

The image depicts a breathtaking landscape bathed in the warm, golden hues of a setting sun. The sky is a dramatic canvas of swirling clouds, painted in shades of pink, orange, and purple, creating a mesmerizing backdrop. The lush green meadow stretches out, dotted with vibrant wildflowers swaying gently in the breeze. Towering trees, their leaves tinged with the soft glow of the sun, stand sentinel along the winding dirt path that meanders through the scene. The overall atmosphere is serene and idyllic, capturing the tranquil beauty of nature at its finest.

05.愤怒嘶吼

A man with tousled dark hair stands in a dramatic landscape, his eyes blazing with fury as he surveys the chaotic scene around him. Clad in a rugged leather jacket, he turns slightly, revealing a determined posture amid a backdrop of crumbling mountains and a valley littered with abandoned structures and scattered flags. The sky is overcast, adding a somber tone to the atmosphere, accentuating his emotional intensity. The camera captures a medium shot, focusing on his tense expression and the desolation surrounding him. The visual style is cinematic with high contrast, enhancing the grim and powerful mood of the moment.

05.旋转灯塔

In the heart of a frozen landscape, a majestic lighthouse stands tall, its stone walls blanketed in a thick layer of snow. The lighthouse, adorned with icicles, emanates a warm, golden glow from its windows, contrasting beautifully with the ethereal green and purple hues of the Northern Lights dancing across the night sky. The icy waters surrounding the structure are dotted with jagged ice formations, creating a surreal and otherworldly scene. The lighthouse keeper's footsteps, lightly imprinted in the snow, lead up the steps to the welcoming wooden door, hinting at the warmth and safety within. The serene, almost magical atmosphere is palpable, as if time itself has slowed to admire this breathtaking winter wonderland.

CogVideoX-5b-I2V镜头控制loras网盘获取:关注公众号口令【CogVideoX-5b-I2V-loras】获取。
更多推荐文章:
• 阿里InContextLoRA:更强ID一致性!基于黑森林F1身份一致性连贯视频分镜图集,10组风格无限创意
• Flux-NewReality:栩栩如生摄影级解禁模型,追求真实细节&风景&神话高品质艺术
• [ComfyUI]InstantIR:来自小红书团队模糊图像修复技术,效果是否惊艳?
• 更像了!5个百分点提升,字节写真换脸PuLID-F1再升级,小红书流量密码
• OmniGen:统一图像生成和多任务集成模型,任意人物自由合影,8位量化体验
• [ComfyUI]Flux:F.1多区域精确控图,无需LORA技术多区域自由构图工具
• [ComfyUI]MochiEdit:最新视频编辑工具,Mochi视频生成加速方案
• [ComfyUI]FaceAging:太好玩啦!仅需几秒看完你或她的一生,从出生到百岁面容
• [ComfyUI]Flux:低显存救星,无限创意!无需部署就能体验最新Joy2|PuLID|LLM等,CF无缝集成
• [ComfyUI]Flux:超治愈!民间青草编织手工艺术,顽强生命微观世界

    感兴趣加入[AGI技术交流群]+V

    如果觉得文章不错,就请在看转发三连

破狼
关注AIGC、LLM、绘图作品、软件工程、技术学习。交流+V:shunshizhiwu。
 最新文章