Thank you very much for your excellent work! Can this model produce anime-style videos?

[0Tzero]

No description provided.

[SHYuanBest]

Thanks for your interest, I think it should work for producing anime-style videos, and I will sample same videos later.

[SHYuanBest]

Something like these?

Videos:

1. https://huggingface.co/datasets/huggingface/documentation-images/resolve/refs%2Fpr%2F406/diffusers/consisid/consisid_output_0.gif

2. https://huggingface.co/datasets/huggingface/documentation-images/resolve/refs%2Fpr%2F406/diffusers/consisid/consisid_output_1.gif

3. https://huggingface.co/datasets/huggingface/documentation-images/resolve/refs%2Fpr%2F406/diffusers/consisid/consisid_output_2.gif

Face Image	Video	Description
		The video, in a beautifully crafted animated style, features a confident woman riding a horse through a lush forest clearing. Her expression is focused yet serene as she adjusts her wide-brimmed hat with a practiced hand. She wears a flowy bohemian dress ......
		The video, in a captivating animated style, shows a woman standing in the center of a snowy forest, her eyes narrowed in concentration as she extends her hand forward. She is dressed in a deep blue cloak, her breath visible in the cold air ......
		The animation features a whimsical portrait of a balloon seller standing in a gentle breeze, captured with soft, hazy brushstrokes that evoke the feel of a serene spring day. His face is framed by a gentle smile, his eyes ......

[0Tzero]

\

Thanks for your interest, I think it should work for producing anime-style videos, and I will sample same videos later.

Thank you very much for your prompt reply! You guys did a great job! However, I tried to use your evaluation method, and found that if the image prompt is an image of an animal, the evaluation method will show that the face cannot be detected and an error is generated, can I use the image prompt of the animal to generate a video?

[SHYuanBest]

Sorry, ConsisID currently only supports the generation of human-like id.