Selfie-With-Younger-Self
| Property | Value |
|---|---|
| Base Model | Wan2.1 14B I2V 480p |
| Training Epochs | 35 |
| Model Format | LoRA Safetensors |
| Model URL | https://huggingface.co/Remade-AI/Selfie-With-Younger-Self |
What is Selfie-With-Younger-Self?
Selfie-With-Younger-Self is an innovative LoRA model trained on the Wan2.1 14B I2V 480p base model that enables the creation of unique videos where individuals appear to take selfies with their younger selves. The model specializes in maintaining consistency in facial features, eye color, and other physical characteristics between the present and younger versions of the subject.
Implementation Details
The model was trained on 40 seconds of carefully curated video data, comprising 9 short clips of people taking selfies with their younger selves. Each clip was individually captioned to ensure precise training. The implementation uses specific trigger words and recommended settings for optimal results.
- LoRA Strength: 1.0
- Embedded Guidance Scale: 6.0
- Flow Shift: 5.0
- Trigger phrase: "s31lf13 taking a selfie with their younger self"
Core Capabilities
- Generation of temporally consistent videos showing subjects with their younger selves
- Maintenance of facial features and characteristics across age variations
- Support for diverse subjects with different physical attributes
- Compatible with ComfyUI workflow through modified Wan Video Wrapper
Frequently Asked Questions
Q: What makes this model unique?
This model specifically focuses on creating realistic interactions between present and past versions of the same person, maintaining consistency in physical features while generating natural-looking selfie scenarios.
Q: What are the recommended use cases?
The model is ideal for creating nostalgic content, personal transformations, and creative visual storytelling where users want to visualize interactions with their younger selves. It works particularly well with the provided prompt template and ComfyUI workflow.
