Loading video...

Video Failed to Load

Go Home

Hi3DGen ๐Ÿ”ฅ High-fidelity 3D geometry generation from a single image by leveraging normal maps as an intermediate representation

Gradio's profile picture

Gradio

54,561 subscribers

50,996 views โ€ข 1 year ago โ€ขvia X (Twitter)

EducationScience & TechnologyArts
LIVE

Anya Rossiโ€ข Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Introducing ๐Ÿ“ฆ๐—”๐—ฟ๐˜๐—ถ๐—Ÿ๐—ฎ๐˜๐—ฒ๐—ป๐˜๐Ÿ”ง (SIGGRAPH Asia 2025) โ€” a high-quality 3D diffusion model that explicitly models object articulation, paving the way for richer, more realistic assets in embodied AI and simulation: โ€“ Generates fully articulated 3D objects โ€“ Physically plausible joints & motion โ€“ High-fidelity 3D Gaussian appearance โ€“ Supports generation from a single real image arXiv: Project: Code (coming soon):
0:16

Introducing ๐Ÿ“ฆ๐—”๐—ฟ๐˜๐—ถ๐—Ÿ๐—ฎ๐˜๐—ฒ๐—ป๐˜๐Ÿ”ง (SIGGRAPH Asia 2025) โ€” a high-quality 3D diffusion model that explicitly models object articulation, paving the way for richer, more realistic assets in embodied AI and simulation: โ€“ Generates fully articulated 3D objects โ€“ Physically plausible joints & motion โ€“ High-fidelity 3D Gaussian appearance โ€“ Supports generation from a single real image arXiv: Project: Code (coming soon):

Xingang Pan

11,473 views โ€ข 7 months ago

Hereโ€™s how to create this cool 3D scene from a single image! ๐Ÿ“ธ Midjourney (isometric image generation) Trellis (Image to 3D Gaussian Splat) Browser Lab (3D Editor Splat Import)
0:13

Hereโ€™s how to create this cool 3D scene from a single image! ๐Ÿ“ธ Midjourney (isometric image generation) Trellis (Image to 3D Gaussian Splat) Browser Lab (3D Editor Splat Import)

Ian Curtis

117,646 views โ€ข 1 year ago

An image + a detailed normal map generated by GeoWizard --> a relightable image. Geowizard is a generative geometry estimation model that could jointly produce high-quality depth and normal with intricate details. Try our huggingface demo here: #huggingface #gradio #geowizard
0:13

An image + a detailed normal map generated by GeoWizard --> a relightable image. Geowizard is a generative geometry estimation model that could jointly produce high-quality depth and normal with intricate details. Try our huggingface demo here: #huggingface #gradio #geowizard

Xiao-Xiao Long

30,338 views โ€ข 2 years ago

3D single-cell maps of embryos reveal "universal" geometry of blood vessels and nerves. More here:
0:17

3D single-cell maps of embryos reveal "universal" geometry of blood vessels and nerves. More here:

Denis Wirtz

35,722 views โ€ข 2 months ago

This is RICULOUSLY good, TRELLIS 3D Generation model by Microsoft! ๐Ÿ”ฅ Generate high-quality 3D assets from text or image prompts. Supports various formats like Radiance Fields, 3D Gaussians, and meshes Available for FREE on Hugging Face!
0:14

This is RICULOUSLY good, TRELLIS 3D Generation model by Microsoft! ๐Ÿ”ฅ Generate high-quality 3D assets from text or image prompts. Supports various formats like Radiance Fields, 3D Gaussians, and meshes Available for FREE on Hugging Face!

Vaibhav (VB) Srivastav

62,437 views โ€ข 1 year ago

Tencent Hunyuan 3D 3.1 is now on Replicate Create 3D models with high texture fidelity and geometry precision with support for up to eight input views.
0:15

Tencent Hunyuan 3D 3.1 is now on Replicate Create 3D models with high texture fidelity and geometry precision with support for up to eight input views.

Replicate

48,529 views โ€ข 4 months ago

๐Ÿš€ Introducing GenLit โ€“ Reformulating Single-Image Relighting as Video Generation! We leverage video diffusion models to perform realistic near-field relighting from just a single imageโ€”No explicit 3D reconstruction or ray tracing required! No intermediate graphics buffers, directly in the pixel space! ๐Ÿ“„ Dive into the paper: ๐ŸŽฅ Project page & demos: ๐Ÿ›  Code coming soon! #GenerativeAI #ComputerVision #Relighting #DiffusionModels #Graphics ๐Ÿงต 1/5
0:19

๐Ÿš€ Introducing GenLit โ€“ Reformulating Single-Image Relighting as Video Generation! We leverage video diffusion models to perform realistic near-field relighting from just a single imageโ€”No explicit 3D reconstruction or ray tracing required! No intermediate graphics buffers, directly in the pixel space! ๐Ÿ“„ Dive into the paper: ๐ŸŽฅ Project page & demos: ๐Ÿ›  Code coming soon! #GenerativeAI #ComputerVision #Relighting #DiffusionModels #Graphics ๐Ÿงต 1/5

Haven Feng @ CVPR

22,427 views โ€ข 1 year ago

๐Ÿคฏ ๐’๐ฅ๐ข๐œ๐ž๐๐ข๐ญ : A revolutionary approach to zero-shot video editingโœ๏ธ๐ŸŽฅ using text-to-image diffusion models ๐ŸŽž๏ธSlicedit achieves great temporal consistency by leveraging spatio-temporal "slices"๐Ÿ”€
0:10

๐Ÿคฏ ๐’๐ฅ๐ข๐œ๐ž๐๐ข๐ญ : A revolutionary approach to zero-shot video editingโœ๏ธ๐ŸŽฅ using text-to-image diffusion models ๐ŸŽž๏ธSlicedit achieves great temporal consistency by leveraging spatio-temporal "slices"๐Ÿ”€

Gradio

19,584 views โ€ข 2 years ago

๐Ÿ˜€ ToonCrafter: Generative Cartoon Interpolation ๐Ÿ”† Can interpolate two cartoon images by leveraging the pre-trained image-to-video diffusion priors. Model on ๐Ÿค—๐Ÿ‘‡
0:10

๐Ÿ˜€ ToonCrafter: Generative Cartoon Interpolation ๐Ÿ”† Can interpolate two cartoon images by leveraging the pre-trained image-to-video diffusion priors. Model on ๐Ÿค—๐Ÿ‘‡

Gradio

93,074 views โ€ข 2 years ago

๐Ÿ†• Image super-resolution model just dropped! Superior results even with a single sampling step. ๐Ÿ”ฅInvSR: Arbitrary-steps Image Super-resolution via Diffusion Inversion.
0:12

๐Ÿ†• Image super-resolution model just dropped! Superior results even with a single sampling step. ๐Ÿ”ฅInvSR: Arbitrary-steps Image Super-resolution via Diffusion Inversion.

Gradio

53,331 views โ€ข 1 year ago

Wonderland: Navigating 3D Scenes from a Single Image Contributions: โ€ข First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. โ€ข Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. โ€ข Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasksโ€”and bridging image space and 3D spaceโ€”through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256ร— spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.
0:12

Wonderland: Navigating 3D Scenes from a Single Image Contributions: โ€ข First, we introduce a representation for controllable 3D generation by leveraging the generative priors from camera-guided video diffusion models. Unlike image models, video diffusion models are trained on extensive video datasets. This enables them to capture comprehensive spatial relationships within scenes across multiple views and embed a form of "3D awareness" in their latent space, which allows us to maintain 3D consistency in novel view synthesis. โ€ข Second, to achieve controllable novel view generation, we empower video models with precise control over specified camera motions. We introduce a novel dual-branch conditioning mechanism that effectively incorporates desired diverse camera trajectories into the video diffusion model. This enables expansion of a single image into a multi-view consistent capture of a 3D scene with precise pose control. โ€ข Third, to achieve efficient 3D reconstruction, we directly transform video latents into 3DGS. We propose a novel latent-based large reconstruction model (LaLRM) that lifts video latents to 3D in a feed-forward manner. With this design, during inference, our model directly predicts 3DGS from a single input image, effectively aligning the generation and reconstruction tasksโ€”and bridging image space and 3D spaceโ€”through the video latent space. Compared with reconstructing scenes from images, the video latent space offers a 256ร— spatial-temporal reduction while retaining essential and consistent 3D structural details. Such a high degree of compression is crucial, as it allows the LaLRM to handle a wider range of 3D scenes within the reconstruction framework, with the same memory constraints.

MrNeRF

52,801 views โ€ข 1 year ago

DreamCraft3D++ can generate animatable high-quality 3D assets from an image in 10 minutes! Yes, it supports meshes ๐Ÿ”ฅ Links โฌ‡๏ธ
0:16

DreamCraft3D++ can generate animatable high-quality 3D assets from an image in 10 minutes! Yes, it supports meshes ๐Ÿ”ฅ Links โฌ‡๏ธ

Dreaming Tulpa ๐Ÿฅ“๐Ÿ‘‘

59,017 views โ€ข 1 year ago

(1/2) ๐Ÿ“ข๐๐๐†๐€: ๐๐ž๐ฎ๐ซ๐š๐ฅ ๐๐š๐ซ๐š๐ฆ๐ž๐ญ๐ซ๐ข๐œ ๐†๐š๐ฎ๐ฌ๐ฌ๐ข๐š๐ง ๐€๐ฏ๐š๐ญ๐š๐ซ๐ฌ ๐Ÿ“ข #SIGGRAPHAsia We leverage a neural parametric representation to facilitate precise control over 3D Gaussians to obtain high-fidelity avatars.
0:19

(1/2) ๐Ÿ“ข๐๐๐†๐€: ๐๐ž๐ฎ๐ซ๐š๐ฅ ๐๐š๐ซ๐š๐ฆ๐ž๐ญ๐ซ๐ข๐œ ๐†๐š๐ฎ๐ฌ๐ฌ๐ข๐š๐ง ๐€๐ฏ๐š๐ญ๐š๐ซ๐ฌ ๐Ÿ“ข #SIGGRAPHAsia We leverage a neural parametric representation to facilitate precise control over 3D Gaussians to obtain high-fidelity avatars.

Matthias Niessner

30,741 views โ€ข 1 year ago

Spell AI making a 3d volume from a single image. Link below
0:13

Spell AI making a 3d volume from a single image. Link below

Spline

10,815 views โ€ข 1 year ago

Leffa from Meta is a ๐Ÿ†• unified framework for controllable person image generation that enables precise manipulation of both appearance (i.e., Virtual try-on) and pose (i.e., Pose transfer). Learn more about ๐Ÿ“š Code,๐Ÿ”ฅ Demo, and ๐Ÿค— Model โฌ‡๏ธ
0:11

Leffa from Meta is a ๐Ÿ†• unified framework for controllable person image generation that enables precise manipulation of both appearance (i.e., Virtual try-on) and pose (i.e., Pose transfer). Learn more about ๐Ÿ“š Code,๐Ÿ”ฅ Demo, and ๐Ÿค— Model โฌ‡๏ธ

Gradio

52,913 views โ€ข 1 year ago

๐Ÿ†• ๐Ÿ”ฅ AniDoc: Animation Creation Made Easier It can colorize a sequence of sketches based on a character design reference with high fidelity, even when the sketches significantly differ in pose and scale. Explore more โฌ‡๏ธ
0:15

๐Ÿ†• ๐Ÿ”ฅ AniDoc: Animation Creation Made Easier It can colorize a sequence of sketches based on a character design reference with high fidelity, even when the sketches significantly differ in pose and scale. Explore more โฌ‡๏ธ

Gradio

37,989 views โ€ข 1 year ago

HOLY SHIT - generate 3D mesh from a single image in LESS THAN A SECOND ๐Ÿคฏ
0:13

HOLY SHIT - generate 3D mesh from a single image in LESS THAN A SECOND ๐Ÿคฏ

Vaibhav (VB) Srivastav

154,108 views โ€ข 1 year ago

Relightify: Relightable 3D Faces from a Single Image via Diffusion Models abs: paper page:
0:10

Relightify: Relightable 3D Faces from a Single Image via Diffusion Models abs: paper page:

AK

85,893 views โ€ข 3 years ago