如今，为自己制作数字形象的方法有很多。我们可以通过 AI 换脸，Midjourney 的角色参考（--cref)，还可以训练自己专属的 LoRA 模型。今天推荐大家一款操作最简单，效果却很出色的工具：RenderNet. 著名 AIGC 达人 TechHalla 分享了用自己的一张照片制作各种数字形象的工作流。
工作流
拍摄一张自己的照片
上传照片到 RenderNet 并起一个名字
选择适当的模型类型（真实/风格化）、年龄、性别、发型
RenderNet 只需要一张正面脸部照片，就能生成一个专属自己的 Flux LoRA 模型。
使用自己的模型生成各种数字形象
生成数字形象的提示词与普通的 AI 绘画提示词没有差别。只需在提示词之中添加先前依据自己照片生成的模型名字：[@ yourcharacter]
[@ yourcharacter] as typical Texan wearing a cowboy hat and denim jacket, and black shoes. Desert western village
[@ yourcharacter] as a green alien | wearing organic lovecraftian armor | purple exoplanet with massive flora background
[@ yourcharacter] bald with long beard as a mad scientist at lab | bottles in the background | red noir saturated film
Avant-garde fashion runway shot | [@ yourcharacter] wearing balenciaga floral suit | male medium body shot | flashy, swag
A close-up shot of a bald hobbit [@ yourcharacter] cooking a chicken over a campfire at the top of a mountain | natural lighting | cinematic composition | fantasy landscape in the background | UHD, realistic textures | early morning mist | dramatic skies
A pirate villain [@ yourcharacter] wielding a saber in an aggressive stance, defending an island during an intense battle | cinematic composition | high-contrast lighting with deep shadows | stormy skies and turbulent seas in the background | dramatic and desaturated
Wide shot of [@ yourcharacter] as a wrestler wearing a tshirt with the text “Flux Consistency” | aggro pose and mood | stadium background
还可以把自己新的形象制作成视频
访问地址：rendernet.ai
评论(0)