正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

🩰 #Live3D #Live2D #FeetTracking 🩰 I’m really happy that this modeling commission also gave me the opportunity to create and experiment with a FeetTracking model using Live3D technology. This technique doesn’t require complex image separation, using only front-view leg, sole, and toenail image assets to achieve pseudo-3D movement and... show more

📐Hephaestus📏Live2D匠人魂

4,885 subscribers

361,792 次观看 • 1 个月前 •via X (Twitter)

艺术科学技术 #Live3D #Live2D #FeetTracking

Anya Rossi• Live Now

Private livecam show

41 条评论

Meep 的头像

Meep1 个月前

Using this model would make it so easy to get gross men to send you money, infinity money glitch

Rynna Aeon ⏳✨ Time Witch Vtuber 【VOops!】的头像

Rynna Aeon ⏳✨ Time Witch Vtuber 【VOops!】1 个月前

Imagine green screening the soles of the feet then getting banned

📐Hephaestus📏Live2D匠人魂的头像

📐Hephaestus📏Live2D匠人魂1 个月前

Haha, why!? They're just feet, something everyone is born with. How could that possibly get me banned? 😇

Rynna Aeon ⏳✨ Time Witch Vtuber 【VOops!】的头像

Rynna Aeon ⏳✨ Time Witch Vtuber 【VOops!】1 个月前

Oh, referencing a different person who got banned for free screening her feet.

Rynna Aeon ⏳✨ Time Witch Vtuber 【VOops!】的头像

Rynna Aeon ⏳✨ Time Witch Vtuber 【VOops!】1 个月前

Green*

VTuberFeetConnoisseur 的头像

VTuberFeetConnoisseur1 个月前

goated

📐Hephaestus📏Live2D匠人魂的头像

📐Hephaestus📏Live2D匠人魂1 个月前

✨🐐✨

Kibbles 🐺🪫 的头像

Kibbles 🐺🪫1 个月前

this is fucking crazy i love it

Kirana🥥💰 的头像

Kirana🥥💰1 个月前

this needs to be an asset

📐Hephaestus📏Live2D匠人魂的头像

📐Hephaestus📏Live2D匠人魂1 个月前

《CREDITS》 😈 VTuber： @chiyouboss 📐 Live2D： @Live2Doll 🐙 Artist： @Kraken_is_sexy

Mintzey 🍨🐈 SEPT. 25TH 2.0 DEBUT 的头像

Mintzey 🍨🐈 SEPT. 25TH 2.0 DEBUT1 个月前

FEEEEEEEEEEEET 🦶

Lotus🦇🪷Batty Overlord 【VToes】的头像

Lotus🦇🪷Batty Overlord 【VToes】1 个月前

Please make this purchasable I NEED ITTTT

Taro 💜⬛ @ setting up my Vgen 💦 的头像

Taro 💜⬛ @ setting up my Vgen 💦1 个月前

the flexible toe gripping and almost typing motions are sending me. this is so well done, but I can't stop staring in disbelief and awe at the power of L2D HAHAH

Podge 💜🥒 『 AURORA 』➝ VeXpo 🇨🇦 的头像

Podge 💜🥒 『 AURORA 』➝ VeXpo 🇨🇦1 个月前

I want one

Oka 的头像

Oka1 个月前

Pour toi @hosarny 🙏

Axbrii ! 🩵☁️【 VTUBER 】的头像

Axbrii ! 🩵☁️【 VTUBER 】1 个月前

IM CRYING I NEED THIS SO I CAN STREAM WITH MY VTUBERS TOES USING MY OVERLAYS AS A LEG REST LMAO

Lillie ♡ Goofy Feettuber 的头像

Lillie ♡ Goofy Feettuber1 个月前

OH MY GOD ? ! ? !

Nakoyama (Commissions Open) 的头像

Nakoyama (Commissions Open)1 个月前

Nobody was supposed to see this @PupunaMai is this the next evolution of the ultimate lifeform, caught in 4K 🤣

Akihiro Tokisaki😈⛓️【lucid ☁️】的头像

Akihiro Tokisaki😈⛓️【lucid ☁️】1 个月前

@neiomicielo Ummmmmm

Gucci 🌿📋 的头像

Gucci 🌿📋1 个月前

Oh yeah

kamo 🍎🌱🪡Debut 2027 . . 的头像

kamo 🍎🌱🪡Debut 2027 . .1 个月前

@maraevantree

FarWestFox🦊🤠 的头像

FarWestFox🦊🤠1 个月前

@zentreya .........don't even deny it LMAO

LARGENAP 昼寝 👁️🪽TTV | Sleep Seraphim 的头像

LARGENAP 昼寝 👁️🪽TTV | Sleep Seraphim1 个月前

Oh my god?? 💀🤣

Meep 的头像

Meep1 个月前

What if you do this pose 🤙 or 🤟

📐Hephaestus📏Live2D匠人魂的头像

📐Hephaestus📏Live2D匠人魂1 个月前

I could give it a try, but the problem is...... I'm not sure how to properly make either of those signs with feet 🤣

UltimateAHP 的头像

UltimateAHP1 个月前

@ZensMeatSlab 😶

kits 🔞 bratty feettuber 的头像

kits 🔞 bratty feettuber1 个月前

OMG

Lushen Zener | Taking a Break 的头像

Lushen Zener | Taking a Break1 个月前

@VTuberNewsDrop @GucciTheMod

Misfit Script ✂️🗞️Paper Spirit Vtuber (OutcastVT) 的头像

Misfit Script ✂️🗞️Paper Spirit Vtuber (OutcastVT)1 个月前

Oh you don't know the Pandora's box you just opened......

📐Hephaestus📏Live2D匠人魂的头像

📐Hephaestus📏Live2D匠人魂1 个月前

You don't know that I'm the one who created that box 😈

Wisp Serahill 🐨🗡️🐩 的头像

Wisp Serahill 🐨🗡️🐩27 天前

@woofle_train this is perfect for you !

CEJ papá luchon 👓 📕 的头像

CEJ papá luchon 👓 📕1 个月前

@ginetapardo mira, si te haces un vtuber le puedes poner patas xD

Gina Zakii 的头像

Gina Zakii1 个月前

@DyamiiVT

SoulNighte 的头像

SoulNighte1 个月前

This is insane

📐Hephaestus📏Live2D匠人魂的头像

📐Hephaestus📏Live2D匠人魂1 个月前

😇💖

DEKAY 🕸️🩸 ⛧ SpiritVeil ⛧ 的头像

DEKAY 🕸️🩸 ⛧ SpiritVeil ⛧1 个月前

@CerbiBites !!!

希律 Ritsu-tan 🍭⛓️ 的头像

希律 Ritsu-tan 🍭⛓️1 个月前

I need this like Vtuber asset 😊😊🙏🙏

Zeaket 🐍✨🛋️😈💚💫🐉⛺💖💸 的头像

Zeaket 🐍✨🛋️😈💚💫🐉⛺💖💸1 个月前

this is the future

Valephore 🌙♦️ 的头像

Valephore 🌙♦️1 个月前

@NyrethVT

Fiishey 的头像

Fiishey1 个月前

@Chadakyo I heard you need something like this

gwimmi | 的头像

gwimmi |1 个月前

@BattyLotus

相关视频

🔱 #Live3D #Live2D #HandTracking 🔱 This is the highest-spec hand-tracking model I’m currently capable of creating. The model was created based on the Live2D-based 2D Pseudo-3D concept, and for the first time, it incorporates techniques such as colored linework, rim lighting, and independent finger outlines. These new techniques allow the model to achieve a pseudo-3D effect while faithfully preserving the artist’s original art style. Please check out the video for more details✨

🔱 #Live3D #Live2D #HandTracking 🔱 This is the highest-spec hand-tracking model I’m currently capable of creating. The model was created based on the Live2D-based 2D Pseudo-3D concept, and for the first time, it incorporates techniques such as colored linework, rim lighting, and independent finger outlines. These new techniques allow the model to achieve a pseudo-3D effect while faithfully preserving the artist’s original art style. Please check out the video for more details✨

📐Hephaestus📏Live2D匠人魂

83,691 次观看 • 3 个月前

🌚 #Live3D #Live2D #HandTracking 🌝 I’d like to talk specifically about a method for creating camouflage pseudo-outlines within the Live2D-based 2D Pseudo-3D concept. At first glance, the visual effect of this technique may look rather subtle. But when it comes to faithfully preserving an illustrator’s art style, it becomes an essential technical detail. This is especially important for illustrators who care deeply about the color details of their linework. With this technique, they can still take control of the pseudo-outline colors even within Live3D. Creating this kind of camouflage pseudo-outline is not difficult. All it requires is for the illustrator to provide one additional special set of assets. It looks like a shadow duplicate of the original hands, except painted with camouflage-like colors. During pseudo-3D rotation, each camouflage-colored area of the shadow duplicate cleverly corresponds to the outline color of the same part. It may be a little hard to imagine from words alone, so please take a look at the video! I hope this special production technique can also bring some inspiration to everyone. 😇

🌚 #Live3D #Live2D #HandTracking 🌝 I’d like to talk specifically about a method for creating camouflage pseudo-outlines within the Live2D-based 2D Pseudo-3D concept. At first glance, the visual effect of this technique may look rather subtle. But when it comes to faithfully preserving an illustrator’s art style, it becomes an essential technical detail. This is especially important for illustrators who care deeply about the color details of their linework. With this technique, they can still take control of the pseudo-outline colors even within Live3D. Creating this kind of camouflage pseudo-outline is not difficult. All it requires is for the illustrator to provide one additional special set of assets. It looks like a shadow duplicate of the original hands, except painted with camouflage-like colors. During pseudo-3D rotation, each camouflage-colored area of the shadow duplicate cleverly corresponds to the outline color of the same part. It may be a little hard to imagine from words alone, so please take a look at the video! I hope this special production technique can also bring some inspiration to everyone. 😇

📐Hephaestus📏Live2D匠人魂

20,959 次观看 • 3 个月前

I had an early access to GPT-6 Astra and I can say, everyone in the world now has a 3D designer at their fingertips. I gave it an image of a house and asked to create it in 3D with all the details including toys, appliances and furniture. This a full 3D model reconstruction in Blender with geometry that you can manually tweak and run 60fps as a locally rendered "game" on device.

I had an early access to GPT-6 Astra and I can say, everyone in the world now has a 3D designer at their fingertips. I gave it an image of a house and asked to create it in 3D with all the details including toys, appliances and furniture. This a full 3D model reconstruction in Blender with geometry that you can manually tweak and run 60fps as a locally rendered "game" on device.

Tom Krcha

1,442,826 次观看 • 1 个月前

To answer all the questions on how this works, the model is actually an image of a plushie that I split into layers and rigged like normal from there in Live2D. Here's a quick speed paint reference to show how the layers were made in CSP using photo manip strats 🧡

To answer all the questions on how this works, the model is actually an image of a plushie that I split into layers and rigged like normal from there in Live2D. Here's a quick speed paint reference to show how the layers were made in CSP using photo manip strats 🧡

Clover🍀🐮🔜San Japan

104,111 次观看 • 1 年前

Dreamina Seedance 2.0 is by far the best video model I've tried, and Dreamina makes it even better. You can try it in the link below. It's not just about the quality of the output, but about how much control you have over your video's look. Watch my video here. I'm attaching reference images and asking the model to generate a video using them. I can reference each image using the @ symbol to instruct the model which image to use. You can even upload a clip and use its camera movement, styles from an image, and audio vibe from a track. By the way, you can take an existing video and replace, remove, or add elements to it while the model preserves everything else. This is the closest we've gotten to "editing videos like photos".

Dreamina Seedance 2.0 is by far the best video model I've tried, and Dreamina makes it even better. You can try it in the link below. It's not just about the quality of the output, but about how much control you have over your video's look. Watch my video here. I'm attaching reference images and asking the model to generate a video using them. I can reference each image using the @ symbol to instruct the model which image to use. You can even upload a clip and use its camera movement, styles from an image, and audio vibe from a track. By the way, you can take an existing video and replace, remove, or add elements to it while the model preserves everything else. This is the closest we've gotten to "editing videos like photos".

Santiago

46,575 次观看 • 6 个月前

I started using Blender through MCP about two weeks ago, and I quickly realized that you can build almost anything with AI. This model was created using Blender, Hunyuan3D, Gemini, and ChatGPT. Here’s how I did it: I opened Gemini, uploaded an image of the Gundam model, and asked it to generate a clean front-view image. I uploaded that front view to ChatGPT and asked it to generate two additional angles: a back view and a 45-degree front-left view. I went to: I signed up with my email and translated the page into English. Then I opened Image to 3D and selected the multi-image option. You’ll see a diagram of a whale from several angles. Upload each reference image in its corresponding position, such as front, 45-degree front-left, and back. I selected the 1.5M-face option. This produces a very high-poly model, but don’t worry, we’ll fix that next. Once the generation is complete, download the model as a GLB file. From the Hunyuan homepage, open 3D Studio using one of the dropdown menus. Select the topology or retopology tool and upload your GLB. I chose the High setting to preserve as much detail as possible. After a few seconds, the model was retopologized. It kept most of its visual detail while using far fewer polygons. The original head and rifle didn’t look very good, so I generated them separately. I returned to ChatGPT and created dedicated reference images for the Gundam’s head and rifle. I generated each part individually in Hunyuan at the 1.5M setting, then ran both through the same retopology process. Next came the textures. Open the texture section, select the multi-image option, and upload the same reference images according to the whale orientation indicators. However, instead of using the original clean textures, I asked ChatGPT to recreate them with wear, rust stains, scratches, and other surface damage. This gave the Gundam a much older and more authentic appearance. Once every part was textured, I imported everything into Blender. You can ask Codex through MCP to remove the original head and rifle, or you can do it manually. Select the main model and press Tab to enter Edit Mode. Press 3 to enable face selection, then press C to activate Circle Select. Paint over the faces belonging to the original helmet or rifle. You can press X to delete those faces or P to separate them into another object. Then position the newly generated, more detailed head and rifle in their place. I demonstrate this process in one of my older videos. And that’s it. You now have a very cool 3D Gundam model! Afterward, I created the cockpit, separated the model into movable sections, and rigged everything for use in my Three.js game. That process deserves its own tutorial, though. If anyone wants to see it, let me know. Or just ask your AI, I guess. They seem to know everything these days. xD

I started using Blender through MCP about two weeks ago, and I quickly realized that you can build almost anything with AI. This model was created using Blender, Hunyuan3D, Gemini, and ChatGPT. Here’s how I did it: I opened Gemini, uploaded an image of the Gundam model, and asked it to generate a clean front-view image. I uploaded that front view to ChatGPT and asked it to generate two additional angles: a back view and a 45-degree front-left view. I went to: I signed up with my email and translated the page into English. Then I opened Image to 3D and selected the multi-image option. You’ll see a diagram of a whale from several angles. Upload each reference image in its corresponding position, such as front, 45-degree front-left, and back. I selected the 1.5M-face option. This produces a very high-poly model, but don’t worry, we’ll fix that next. Once the generation is complete, download the model as a GLB file. From the Hunyuan homepage, open 3D Studio using one of the dropdown menus. Select the topology or retopology tool and upload your GLB. I chose the High setting to preserve as much detail as possible. After a few seconds, the model was retopologized. It kept most of its visual detail while using far fewer polygons. The original head and rifle didn’t look very good, so I generated them separately. I returned to ChatGPT and created dedicated reference images for the Gundam’s head and rifle. I generated each part individually in Hunyuan at the 1.5M setting, then ran both through the same retopology process. Next came the textures. Open the texture section, select the multi-image option, and upload the same reference images according to the whale orientation indicators. However, instead of using the original clean textures, I asked ChatGPT to recreate them with wear, rust stains, scratches, and other surface damage. This gave the Gundam a much older and more authentic appearance. Once every part was textured, I imported everything into Blender. You can ask Codex through MCP to remove the original head and rifle, or you can do it manually. Select the main model and press Tab to enter Edit Mode. Press 3 to enable face selection, then press C to activate Circle Select. Paint over the faces belonging to the original helmet or rifle. You can press X to delete those faces or P to separate them into another object. Then position the newly generated, more detailed head and rifle in their place. I demonstrate this process in one of my older videos. And that’s it. You now have a very cool 3D Gundam model! Afterward, I created the cockpit, separated the model into movable sections, and rigged everything for use in my Three.js game. That process deserves its own tutorial, though. If anyone wants to see it, let me know. Or just ask your AI, I guess. They seem to know everything these days. xD

Spectro

81,176 次观看 • 2 个月前

This first-person shooter was built from a single prompt, using 404-generated 3D assets. On the latest Novelty Search, 404 explains how SN17 is moving from Titan, its open image-to-3D model built with miners, toward 3D assets generated as code that AI agents can edit and use directly in games. That work is part of a broader direction for 404: building world models where AI agents can learn and act. Hosted by const Full episode in the first comment

This first-person shooter was built from a single prompt, using 404-generated 3D assets. On the latest Novelty Search, 404 explains how SN17 is moving from Titan, its open image-to-3D model built with miners, toward 3D assets generated as code that AI agents can edit and use directly in games. That work is part of a broader direction for 404: building world models where AI agents can learn and act. Hosted by const Full episode in the first comment

Openτensor Foundaτion

19,046 次观看 • 1 个月前

Can GPT-6 Astra actually create a 3D model like this? 🤔 I’ve been seeing people hype GPT-6 Astra looooot when it comes to 3D modeling, but can it really create 3D models with this level of detail? Wish i had access to try it,.. anyway, I created this one using Tripo as my lil submission to Tripo’s Companion Challenge where you have to create a 3D companion of your choice. I generated a bunch of different imaginary companions using GPT Image 2.0, from cute little dragons and humans to these adorable animals. After trying out a lot of different ideas, I finally settled on this panda in a bamboo forest look. I love how quickly you can go from having a random idea in your head to actually seeing it come to life, these days. That part is honestly so much fun. Now I just need to animate this little guy and turn it into an animated panda game or maybe 3D print it out haha #tripop2

Can GPT-6 Astra actually create a 3D model like this? 🤔 I’ve been seeing people hype GPT-6 Astra looooot when it comes to 3D modeling, but can it really create 3D models with this level of detail? Wish i had access to try it,.. anyway, I created this one using Tripo as my lil submission to Tripo’s Companion Challenge where you have to create a 3D companion of your choice. I generated a bunch of different imaginary companions using GPT Image 2.0, from cute little dragons and humans to these adorable animals. After trying out a lot of different ideas, I finally settled on this panda in a bamboo forest look. I love how quickly you can go from having a random idea in your head to actually seeing it come to life, these days. That part is honestly so much fun. Now I just need to animate this little guy and turn it into an animated panda game or maybe 3D print it out haha #tripop2

The Bugged Dev

160,498 次观看 • 1 个月前

Claude Fable 5 has impressive spatial reasoning capabilities that are immediately relevant for engineering and design: In this example I gave Claude Fable 5 a single photo of a hierarchical mesh torus (image generated using a text-to-image model); and in one shot, no iteration, it built a full interactive 3D simulation: ~1,400 nodes & 4,400 fibers, realistic mass–spring physics you can grab, compress, twist… plus strain-based sonification so you can 'hear' the structure vibrate. It inferred the complete 3D topology from a partial 2D view. ⤵️

Claude Fable 5 has impressive spatial reasoning capabilities that are immediately relevant for engineering and design: In this example I gave Claude Fable 5 a single photo of a hierarchical mesh torus (image generated using a text-to-image model); and in one shot, no iteration, it built a full interactive 3D simulation: ~1,400 nodes & 4,400 fibers, realistic mass–spring physics you can grab, compress, twist… plus strain-based sonification so you can 'hear' the structure vibrate. It inferred the complete 3D topology from a partial 2D view. ⤵️

Markus J. Buehler

36,684 次观看 • 3 个月前

This was way too easy... I created a cute 3D character in minutes instead of hours using this combo: Unbound + Tripo I’m not a 3D artist, but this made the whole process feel simple, playful, and actually accessible. Unbound helped me shape the character idea, then Tripo turned it into a 3D model without the usual complicated workflow If you are a 3D artist or you are just starting, try Unbound. It’s free, open for everyone to experiment with, and a fun way to create characters, creatures, and 3D models without feeling overwhelmed I had so much fun using it today, and I think many visual creators are going to love it. Link in the first comment 👇🏻

This was way too easy... I created a cute 3D character in minutes instead of hours using this combo: Unbound + Tripo I’m not a 3D artist, but this made the whole process feel simple, playful, and actually accessible. Unbound helped me shape the character idea, then Tripo turned it into a 3D model without the usual complicated workflow If you are a 3D artist or you are just starting, try Unbound. It’s free, open for everyone to experiment with, and a fun way to create characters, creatures, and 3D models without feeling overwhelmed I had so much fun using it today, and I think many visual creators are going to love it. Link in the first comment 👇🏻

Amira Zairi

186,621 次观看 • 5 个月前

You don't need to be a 3D artist to start building 3D websites anymore 💛 Built this cool 3D jet drone website in Three.js using Claude Opus 5 and Tripo in < 1 hour. A lot of people still think AI is great for generating code, but not for building something that actually feels polished and interactive or "complex". But it has been changing super fast these days. The model you see here is only 946KB, compressed from its original 13.6MB size automatically by Claude without losing any quality. This kind of intelligence has been developed in many other AI tools for anything related to 3D (except GPT 5.6 sol lol). You just have to describe what you want and they handle everything "engineerly" (that's not even a word but okay :D) Anyway, also you don't really have to learn anything about 3D modeling to create these models. All you need is a good image and put it in Tripo. For this case, i generated model using "smart mesh" feature that generated a cleaner low-poly topology and edge flow. That helps keep your actual 3D asset even more lightweight. Earlier, with HD model feature, my model size with 8k texture was around 60-70MB but with Smart Mesh, it's around 13-16MB with 8k texture applied separately. Of course, this isn't something I knew from the beginning. All I did was "try" out different features and see what happens. Live: Code: Design:

You don't need to be a 3D artist to start building 3D websites anymore 💛 Built this cool 3D jet drone website in Three.js using Claude Opus 5 and Tripo in < 1 hour. A lot of people still think AI is great for generating code, but not for building something that actually feels polished and interactive or "complex". But it has been changing super fast these days. The model you see here is only 946KB, compressed from its original 13.6MB size automatically by Claude without losing any quality. This kind of intelligence has been developed in many other AI tools for anything related to 3D (except GPT 5.6 sol lol). You just have to describe what you want and they handle everything "engineerly" (that's not even a word but okay :D) Anyway, also you don't really have to learn anything about 3D modeling to create these models. All you need is a good image and put it in Tripo. For this case, i generated model using "smart mesh" feature that generated a cleaner low-poly topology and edge flow. That helps keep your actual 3D asset even more lightweight. Earlier, with HD model feature, my model size with 8k texture was around 60-70MB but with Smart Mesh, it's around 13-16MB with 8k texture applied separately. Of course, this isn't something I knew from the beginning. All I did was "try" out different features and see what happens. Live: Code: Design:

The Bugged Dev

51,558 次观看 • 1 个月前

HOLY SHIT! Sparc 3D is INSANE! All they need to do now is enable PBR texture support, object separation and decimation and this in my opinion completely replaces photogrammetry. The quality of the reconstruction of these 3D objects from a single image is mind blowing to me. Yes, the model is insanely dense to allow all of this detail but the fact that AI can generate with this level of fidelity is CRAZY!!! You can try it for free on Huggingface

HOLY SHIT! Sparc 3D is INSANE! All they need to do now is enable PBR texture support, object separation and decimation and this in my opinion completely replaces photogrammetry. The quality of the reconstruction of these 3D objects from a single image is mind blowing to me. Yes, the model is insanely dense to allow all of this detail but the fact that AI can generate with this level of fidelity is CRAZY!!! You can try it for free on Huggingface

Travis Davids

18,230 次观看 • 1 年前

How to Turn Any Live2D Model into a GIFtuber! If VTube Studio struggles on your PC during streams, this is a lighter alternative you can try. It is a bit of a process, but the final result is worth it for me. I tried to keep the video as simple and condensed as possible, do pause if you need to! I also played around more with customising my Akizone Prism Model and I am so happy with it now 🥰

How to Turn Any Live2D Model into a GIFtuber! If VTube Studio struggles on your PC during streams, this is a lighter alternative you can try. It is a bit of a process, but the final result is worth it for me. I tried to keep the video as simple and condensed as possible, do pause if you need to! I also played around more with customising my Akizone Prism Model and I am so happy with it now 🥰

✘ sony

17,685 次观看 • 8 个月前

📢📢 𝐀𝐯𝐚𝐭𝟑𝐫 📢📢 Avat3r creates high-quality 3D head avatars from just a few input images in a single forward pass with a new dynamic 3DGS reconstruction model. Video: Project: Our core idea is to make Gaussian Reconstruction Models animatable. We find that a simple cross-attention to an expression code sequence is already sufficient to model complex facial expressions. We then incorporate position maps from DUSt3R and feature maps from Sapiens to facilitate the prediction task. While DUSt3R's position maps act as a pixel-aligned initialization for the Gaussians' positions, the Sapiens feature maps help the cross-view transformer to match corresponding image tokens in the 4 input images. One major challenge in creating a 3D head avatar from smartphone images comes from inconsistent facial expressions when the subject could not remain perfectly static during the capture. We eliminate this static requirement by simply showing our model input images with different facial expressions during training. This technique makes our model robust to inconsistent input images later on. Finally, we show that despite the model has been trained with 4 input images, one can even create a 3D head avatar when only a single image is available. To achieve this, we employ a pre-trained 3D GAN to lift the single image to 3D and then render the 4 input images for our model. This allows us to create 3D head avatars from single images and even highly out-of-distribution examples like AI generated faces, paintings or statues. Great work by Tobias Kirschstein from his internship at Meta with Javier Romero, Artem Sevastopolsky, and Shunsuke Saito

Matthias Niessner

74,863 次观看 • 1 年前

Here are some tips for those who want to practice face acting: ➡️U motion: Move your head in a "U" Shape so it's never stuck in the same place. This will help your movements look more interesting. ➡️Key points: When practicing with your model, learn to get to a "Key point." This is when you can automatically get to an expression without using toggles. I use surprised and angry expressions a lot, so I know exactly how to move my face to achieve that look, mainly for clips or comedic effect. I practiced enough so I know what it will look like on the model. ➡️Head tilts: If you struggle with face acting, a simple head tilt will go a long way. Overall, the movement and face acting of your model does matter. It can help it look more interesting, but also, it's okay if you can't face acting all the time. Using just key animations when you're gaming will work perfectly. Example: You died in a game, and you go to the key point of surprise. You know, making this face will make your model look shocked, so you will do a dramatic eye open and mouth agape to achieve this, to reflect onto your model. Lastly. The way your model moves is usually 30% rigging artist,20% tracking (webcam/iPhone), and 50% you. Your rigging artist can give you the tools, but only you can make your models' results reflect your emotions. As always, DO YOUR PARAMETERS AND SETTINGS IN VTUBE STUDIO. It will not be perfect to your face and camera settings unless you adjust the model, (This is an old video I did on parameters) Holy Yapfest.

Here are some tips for those who want to practice face acting: ➡️U motion: Move your head in a "U" Shape so it's never stuck in the same place. This will help your movements look more interesting. ➡️Key points: When practicing with your model, learn to get to a "Key point." This is when you can automatically get to an expression without using toggles. I use surprised and angry expressions a lot, so I know exactly how to move my face to achieve that look, mainly for clips or comedic effect. I practiced enough so I know what it will look like on the model. ➡️Head tilts: If you struggle with face acting, a simple head tilt will go a long way. Overall, the movement and face acting of your model does matter. It can help it look more interesting, but also, it's okay if you can't face acting all the time. Using just key animations when you're gaming will work perfectly. Example: You died in a game, and you go to the key point of surprise. You know, making this face will make your model look shocked, so you will do a dramatic eye open and mouth agape to achieve this, to reflect onto your model. Lastly. The way your model moves is usually 30% rigging artist,20% tracking (webcam/iPhone), and 50% you. Your rigging artist can give you the tools, but only you can make your models' results reflect your emotions. As always, DO YOUR PARAMETERS AND SETTINGS IN VTUBE STUDIO. It will not be perfect to your face and camera settings unless you adjust the model, (This is an old video I did on parameters) Holy Yapfest.

♥️Cutiepierin | L2D RIG ARTIST

94,177 次观看 • 11 个月前