In today’s video, we’re diving into a groundbreaking development in AI video technology—character referencing in text-to-video models. For the first time, I’m seeing this feature implemented, and it’s coming from an unexpected source.

I’ll walk you through the three major workflows for AI video creation—text-to-video, image-to-video, and video-to-video—and discuss their strengths and weaknesses. You’ll see examples of both the best and worst-case scenarios, along with tips and tricks I’ve learned along the way.

We’ll explore how Video allows you to upload a reference image to create a consistent character in your AI-generated videos, a feature that could potentially set it apart from other models. Whether you’re into generating cinematic scenes or just want to experiment with AI video creation, this video is packed with insights and practical advice.

LINKS:
Vidu: [ Ссылка ]
Krea: [ Ссылка ]
Topaz: [ Ссылка ]

CHAPTERS
00:00 - Introduction
00:21 - Overview of AI Video Workflows
00:47 - Strengths and Weaknesses of Text-to-Video
1:37 - Vidu's New Feature
2:25 - Vidu Overview
03:38 - Example: Character Consistency Tests
4:28.- Where Character Referencing doesn't Work
5:09 Our Pirate Lass
5:57 Using Real Photos is remarkable
07:12 - Upscaling Needed?
7:52 Krea's secret weapon
09:00 Community Outputs
9:40 Is Vidu a Killer?
10:09: My Bone to Pick

свернуть