Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Meta presents Garment3DGen! A method that can stylize the geometry and textures from 2D image and 3D mesh garments, which can be fitted on top of parametric bodies and be simulated ๐Ÿ”ฅ More examples โฌ‡๏ธ

dreamingtulpa's profile picture

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

54,765 subscribers

157,969 Aufrufe โ€ข vor 2 Jahren โ€ขvia X (Twitter)

Wissenschaft & Technologie
LIVE

Anya Rossiโ€ข Live Now

Private livecam show

10 Kommentare

Profilbild von Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘
Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘vor 2 Jahren

The method supports mesh deformations like turning a tank-top into a dress.

Profilbild von Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘
Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘vor 2 Jahren

ControlNet enables turning rough 2D sketches into images which can then be turned into animatable 3D garments using Garment3DGen.

Profilbild von Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘
Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘vor 2 Jahren

Bonus: Knight armor

Profilbild von Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘
Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘vor 2 Jahren

The code isn't available yet, but luckily it's from Meta, so there is a chance this will drop soon! In the mean time, you can learn more about Garment3DGen here: For more Generative AI fun: โœจ Follow yours truly @dreamingtulpa โœจ Like and repost ๐Ÿ‘‡

Profilbild von Tyrell Corporation Operations
Tyrell Corporation Operationsvor 2 Jahren

@in3d_nz Oh god, more men in dresses...

Profilbild von Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘
Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘vor 2 Jahren

@in3d_nz ๐Ÿ˜‚

Profilbild von Key
Keyvor 2 Jahren

@naaveelab Love the innovative approach with Garment3DGen, enhancing parametric bodies with stylized 2D image textures - truly pushing the boundaries of fashion technology!

Profilbild von Xiaoyu Xiang
Xiaoyu Xiangvor 2 Jahren

Thanks for sharing our work!

Profilbild von bitpixi ๐Ÿงš๐Ÿปโ€โ™‚๏ธ
bitpixi ๐Ÿงš๐Ÿปโ€โ™‚๏ธvor 2 Jahren

@dankvr

Profilbild von alan
alanvor 2 Jahren

ๆœŸๅพ…(โœงโˆ€โœง)

ร„hnliche Videos

Nvidia presents Edify 3D! The method can generate high-quality 3D assets from text descriptions. It uses a diffusion model to create detailed quad-mesh topologies and high-resolution textures in under 2 minutes.
0:43

Nvidia presents Edify 3D! The method can generate high-quality 3D assets from text descriptions. It uses a diffusion model to create detailed quad-mesh topologies and high-resolution textures in under 2 minutes.

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

39,517 Aufrufe โ€ข vor 1 Jahr

This video looks at converting a Displacement texture into actual geometry, that can be 3D printed or sculpted on. Textures from Polyhaven. #Blender #B3D #Blender3d
0:45

This video looks at converting a Displacement texture into actual geometry, that can be 3D printed or sculpted on. Textures from Polyhaven. #Blender #B3D #Blender3d

Jan van den Hemel

10,549 Aufrufe โ€ข vor 1 Jahr

Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors paper page: present Magic123, a two-stage coarse-to-fine approach for high-quality, textured 3D meshes generation from a single unposed image in the wild using both2D and 3D priors. In the first stage, we optimize a neural radiance field to produce a coarse geometry. In the second stage, we adopt a memory-efficient differentiable mesh representation to yield a high-resolution mesh with a visually appealing texture. In both stages, the 3D content is learned through reference view supervision and novel views guided by a combination of 2D and 3D diffusion priors. We introduce a single trade-off parameter between the 2D and 3D priors to control exploration (more imaginative) and exploitation (more precise) of the generated geometry. Additionally, we employ textual inversion and monocular depth regularization to encourage consistent appearances across views and to prevent degenerate solutions, respectively. Magic123 demonstrates a significant improvement over previous image-to-3D techniques, as validated through extensive experiments on synthetic benchmarks and diverse real-world images.
0:34

Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors paper page: present Magic123, a two-stage coarse-to-fine approach for high-quality, textured 3D meshes generation from a single unposed image in the wild using both2D and 3D priors. In the first stage, we optimize a neural radiance field to produce a coarse geometry. In the second stage, we adopt a memory-efficient differentiable mesh representation to yield a high-resolution mesh with a visually appealing texture. In both stages, the 3D content is learned through reference view supervision and novel views guided by a combination of 2D and 3D diffusion priors. We introduce a single trade-off parameter between the 2D and 3D priors to control exploration (more imaginative) and exploitation (more precise) of the generated geometry. Additionally, we employ textual inversion and monocular depth regularization to encourage consistent appearances across views and to prevent degenerate solutions, respectively. Magic123 demonstrates a significant improvement over previous image-to-3D techniques, as validated through extensive experiments on synthetic benchmarks and diverse real-world images.

AK

305,663 Aufrufe โ€ข vor 3 Jahren

MetaHumans are cool, but they might be soon obsolete! UltrAvatar can generate realistic and animatable 3D avatars with PBR textures from only a text prompt or a single image.
0:31

MetaHumans are cool, but they might be soon obsolete! UltrAvatar can generate realistic and animatable 3D avatars with PBR textures from only a text prompt or a single image.

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

28,152 Aufrufe โ€ข vor 2 Jahren

Nature presents a captivating confluence of similarity and diversity. Our new method 3D-Fauna learns a pan-category articulated 3D model of quadruped animals from Internet photos. At test time, it turns a single image into an animatable textured 3D mesh in a feed-forward pass.
0:22

Nature presents a captivating confluence of similarity and diversity. Our new method 3D-Fauna learns a pan-category articulated 3D model of quadruped animals from Internet photos. At test time, it turns a single image into an animatable textured 3D mesh in a feed-forward pass.

Elliott / Shangzhe Wu

107,286 Aufrufe โ€ข vor 2 Jahren

David Lammy and Labour are passing a law which means that existing graves can be exhumed and the bodies cremated so that they can be re-used by others. So who benefits from this - the religion that insists on burials and no cremations?๐Ÿค”๐Ÿค”๐Ÿค”
3:02

David Lammy and Labour are passing a law which means that existing graves can be exhumed and the bodies cremated so that they can be re-used by others. So who benefits from this - the religion that insists on burials and no cremations?๐Ÿค”๐Ÿค”๐Ÿค”

Dave

243,249 Aufrufe โ€ข vor 3 Monaten

You can build your 3D models directly from a prompt into Claude code. All thanks to this beautiful python project, build123 a parametric modeling framework for 2D and 3D CAD. It is built on the Open Cascade geometric kernel and it is suitable for 3D printing.
0:36

You can build your 3D models directly from a prompt into Claude code. All thanks to this beautiful python project, build123 a parametric modeling framework for 2D and 3D CAD. It is built on the Open Cascade geometric kernel and it is suitable for 3D printing.

Marco Franzon

20,777 Aufrufe โ€ข vor 4 Monaten

Think spatially #echofirst The leaflet orientation of the mechanical mitral valve prosthesis on 3D echocardiography can be predicted from the 2D TEE image. Chittur Sivaram MD MACP MACC
0:45

Think spatially #echofirst The leaflet orientation of the mechanical mitral valve prosthesis on 3D echocardiography can be predicted from the 2D TEE image. Chittur Sivaram MD MACP MACC

Alexander Mladenow MD

12,768 Aufrufe โ€ข vor 1 Monat

AI can synthesize dynamic textures on 3D meshes without UV maps! MeshNCA is a model that can generate dynamic textures from images, text prompts, and motion vector fields.
1:19

AI can synthesize dynamic textures on 3D meshes without UV maps! MeshNCA is a model that can generate dynamic textures from images, text prompts, and motion vector fields.

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

79,593 Aufrufe โ€ข vor 2 Jahren

By popular demand, we're sharing more examples that showcase CSM Cube's ability to generate high-quality 3D mesh topology and distinct parts. The results shown were achieved using our single-click Image-to-3D feature, but users can obtain even better quality with careful prompting and editing.
0:46

By popular demand, we're sharing more examples that showcase CSM Cube's ability to generate high-quality 3D mesh topology and distinct parts. The results shown were achieved using our single-click Image-to-3D feature, but users can obtain even better quality with careful prompting and editing.

Common Sense Machines

16,904 Aufrufe โ€ข vor 11 Monaten

AI can now create interactive 3D worlds from just image and text promptsโ€ฆ Literally be the penguin
0:59

AI can now create interactive 3D worlds from just image and text promptsโ€ฆ Literally be the penguin

Min Choi

62,250 Aufrufe โ€ข vor 5 Monaten

If anyone is needing any vtube work done this week let me know! I can basically take any 3D or 2d image and rig it up to be used for videos and streaming Finally got caught up on the last few. Hereโ€™s some examples of what they look like.
0:20

If anyone is needing any vtube work done this week let me know! I can basically take any 3D or 2d image and rig it up to be used for videos and streaming Finally got caught up on the last few. Hereโ€™s some examples of what they look like.

๐Ÿ”ฅburningbush.eth๐Ÿ”ฅ

19,219 Aufrufe โ€ข vor 2 Jahren

Adobe is entering the image-to-3D game! Their new method, LRM, can create high-fidelity 3D meshes from a single image in just 5 seconds ๐Ÿ”ฅ It's trained on 1 million objects and is able to generate objects from real-world images and generative AI models.
0:28

Adobe is entering the image-to-3D game! Their new method, LRM, can create high-fidelity 3D meshes from a single image in just 5 seconds ๐Ÿ”ฅ It's trained on 1 million objects and is able to generate objects from real-world images and generative AI models.

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

270,942 Aufrufe โ€ข vor 2 Jahren

Nvidia presents EdgeRunner! The method can generate high-quality 3D meshes with up to 4,000 faces at a spatial resolution of 512 from images and point-clouds.
0:29

Nvidia presents EdgeRunner! The method can generate high-quality 3D meshes with up to 4,000 faces at a spatial resolution of 512 from images and point-clouds.

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

173,103 Aufrufe โ€ข vor 1 Jahr

3D Gaussian Splats are cool, but they're static (Part 25). DG-Mesh can reconstruct high-quality and time-consistent 3D meshes from a single video and track mesh vertices over time, which enables texture editing on dynamic objects.
0:22

3D Gaussian Splats are cool, but they're static (Part 25). DG-Mesh can reconstruct high-quality and time-consistent 3D meshes from a single video and track mesh vertices over time, which enables texture editing on dynamic objects.

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

79,033 Aufrufe โ€ข vor 2 Jahren

Technicians spent more than a week scanning and stabilizing โ€œGugusse and the Automaton" โ€“ which features, by the way, the first known moving image of what might be called a robot โ€“ so that it can now be enjoyed by anyone online!
2:59

Technicians spent more than a week scanning and stabilizing โ€œGugusse and the Automaton" โ€“ which features, by the way, the first known moving image of what might be called a robot โ€“ so that it can now be enjoyed by anyone online!

Library of Congress

25,932 Aufrufe โ€ข vor 5 Monaten

๐Ÿ“ข SHeaP: Self-Supervised Head Predictor Learned via 2D Gaussians ๐Ÿ“ข Given a single input image, we predict accurate 3D head geometry, pose, and expression. Previous works (e.g. DECA, EMOCA) use differentiable mesh rasterization to learn a self-supervised head geometry predictor via a photometric reconstruction loss. We borrow these ideas, but our key insight is to replace the mesh rendering with 2D Gaussian Splatting. This leads to much higher accuracy of the underlying predicted geometry and thus more gradient signal during training. ๐ŸŒ ๐ŸŽฅ Great work by Liam Schoneveld Davide Davoli Jiapeng Tang
0:48

๐Ÿ“ข SHeaP: Self-Supervised Head Predictor Learned via 2D Gaussians ๐Ÿ“ข Given a single input image, we predict accurate 3D head geometry, pose, and expression. Previous works (e.g. DECA, EMOCA) use differentiable mesh rasterization to learn a self-supervised head geometry predictor via a photometric reconstruction loss. We borrow these ideas, but our key insight is to replace the mesh rendering with 2D Gaussian Splatting. This leads to much higher accuracy of the underlying predicted geometry and thus more gradient signal during training. ๐ŸŒ ๐ŸŽฅ Great work by Liam Schoneveld Davide Davoli Jiapeng Tang

Matthias Niessner

28,559 Aufrufe โ€ข vor 1 Jahr

Adobe presents SynCHMR! A method that can reconstruct camera trajectory, human motion, and scene in one global coordinate from videos ๐Ÿ”ฅ
0:26

Adobe presents SynCHMR! A method that can reconstruct camera trajectory, human motion, and scene in one global coordinate from videos ๐Ÿ”ฅ

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

108,628 Aufrufe โ€ข vor 2 Jahren

MVDream: Multi-view Diffusion for 3D Generation paper page: propose MVDream, a multi-view diffusion model that is able to generate geometrically consistent multi-view images from a given text prompt. By leveraging image diffusion models pre-trained on large-scale web datasets and a multi-view dataset rendered from 3D assets, the resulting multi-view diffusion model can achieve both the generalizability of 2D diffusion and the consistency of 3D data. Such a model can thus be applied as a multi-view prior for 3D generation via Score Distillation Sampling, where it greatly improves the stability of existing 2D-lifting methods by solving the 3D consistency problem. Finally, we show that the multi-view diffusion model can also be fine-tuned under a few shot setting for personalized 3D generation, i.e. DreamBooth3D application, where the consistency can be maintained after learning the subject identity.
0:44

MVDream: Multi-view Diffusion for 3D Generation paper page: propose MVDream, a multi-view diffusion model that is able to generate geometrically consistent multi-view images from a given text prompt. By leveraging image diffusion models pre-trained on large-scale web datasets and a multi-view dataset rendered from 3D assets, the resulting multi-view diffusion model can achieve both the generalizability of 2D diffusion and the consistency of 3D data. Such a model can thus be applied as a multi-view prior for 3D generation via Score Distillation Sampling, where it greatly improves the stability of existing 2D-lifting methods by solving the 3D consistency problem. Finally, we show that the multi-view diffusion model can also be fine-tuned under a few shot setting for personalized 3D generation, i.e. DreamBooth3D application, where the consistency can be maintained after learning the subject identity.

AK

294,496 Aufrufe โ€ข vor 2 Jahren