Moore AnimateAnyone Tutorial: Make photo dance with AI Dance Animator

hello everyone I am UDA akiyama this time it'll show you animate anyone which can make images dance it's been popular since the end of last year it has been a trending topic since the start of last year but until now it was only available on the introduction page and couldn't be used however the reproduction and implementation of animate anyone has finally been completed and now it can be used to create moving images earlier I introduced magic animate a tool that uses AI technology to create videos from still images while magic animate was developed by a company from Tik Tok animate anyone is a tool created by a Chinese IT company called Alibaba let's take a closer look at what's [Music] inside the animation source code has not been made public yet and there is currently no platform to use it on so this time 'll be using animate anyone from this page we can achieve a this by replicating and applying animate anyone just like magic animate which I introduced earlier this one also uses motion synthesis to move still images however animate anyone seems to be a tool that animates using skeletal videos just like a human being as you can see there are various options available there are demo videos available and you can see how accurate the movements are while maintaining consistency you can also see the accuracy of the original video movements including hair and clothing so I am going to actually move animate anyone by clicking here you can control it from the hugging face demo page however sometimes the movement can be quite slow so this time it'll be using Google collab you can find the Google collab Link in the summary section and run it from there click on the link in the summary section and and open it in Google collab you can simply run it by clicking on the play button here let's run it the initial setup for animate anyone is done and the URL for this running screen has been displayed please click this fabric URL when you click on it a screen like this will pop up and start running no changes needed the current translation is already clear and modern [Music] next up it'll be uploading a motion synthesis but if you scroll down you'll find an example so it'll just use that one to use this motion synthesis just click here no changes needed the translation is already contextual and modern this is what I am going to use this time it looks like this if you use the example the reference image will be updated so it'll upload it again next I want to adjust the video's length and size this time I want to use the portrait image size so it'll set the width to 512 and the height to 768 please make sure the video size matches this motion capture video be careful not to accidentally change the size of this reference image next it'll set the video length to 120 which is the longest duration for this one the video will be 4 seconds long so I set the length to 120 the original length was 24 which would have made it a 0.8c video so now it'll actually click on animate yep the processing is done now this time I made a 4-second video with a length of 120 which took around 26 minutes it sure took a while huh if you are planning to make a video it's a good idea to preview a shorter version first to see how it turns out before making the full- length one Yep this is the final video yes here it is yeah can you clearly see the movements when you watch the original video the AI recognizes that the lower part of the knee is missing and adds it even if it's not visible in the image if you take a closer look at their clothes and hair you'll notice that their movements are quite natural however the face still looks a bit distorted so I believe it's still at a practical level I made this using other images and motion synthesis videos but the face still looks a bit distorted by the way the distorted face can be corrected using stable diffusion this video was fixed with stable diffusion this is a face correction technique using a control net that only uses reference images it might be a bit clumsy so if you have any suggestions feel free to drop the in the comments hey guys how did you like the video if you found it helpful please give it a thumbs up and don't forget to subscribe to my channel if you have any thoughts or feedback don't hesitate to drop them in the comments below see you guys in the next video goodbye

Channel: AI Lab Tutorial

Views: 1,441

Keywords: stable diffusion, stable diffusion tutorial, ai animation, AI-generated animations, YouTube short videos, denoising settings, realistic animations, stable diffusion img2img, ai art, install stable diffusion tutorial, move-to-move extension, AI animations, Stable Diffusions, virtual characters, YouTube shorts, dance videos, animation stable diffusion, animation stable diffusion tutorial, magic animate

Id: KVWmyLqk0nw

Length: 5min 17sec (317 seconds)

Published: Sun Mar 17 2024