Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

📢 Lyra: Generative 3D Scene Reconstruction via Video Diffusion Model Self-Distillation Got only one or a few images and wondering if recovering the 3D environment is a reconstruction or generation problem? Why not do it with a generative reconstruction model! We show that a camera-conditioned video diffusion model can... show more

Sherwin Bahmani

1,393 subscribers

66,918 views • 1 year ago •via X (Twitter)

Science & Technology News & Politics Education

Anya Rossi• Live Now

Private livecam show

38 Comments

Sherwin Bahmani1 year ago

Previous work (left) trains image-to-3D models using real-world datasets with limited diversity. In contrast, we propose a self-distillation framework (right) for generative 3D scene reconstruction. Specifically, a pre-trained camera-controlled video diffusion model (GEN3C) with its RGB decoder output (teacher) supervises the rendering of the 3DGS decoder (student). Using a video model pre-trained on large-scale real-world 2D video data allows us to provide diverse and potentially infinite multi-view supervision.

Sherwin Bahmani1 year ago

Our model generates 3DGS scenes from a single input image in a feed-forward manner generalizing across a wide range of scenarios., without the need for per-scene optimization.

Sherwin Bahmani1 year ago

Moreover, our framework easily extends to feed-forward 4D scene generation from a single input video.

Sherwin Bahmani1 year ago

Our approach seamlessly generalizes to full 3D scene generation from a single real image, without ever being trained on real-world data.

Sherwin Bahmani1 year ago

The generated 3D Gaussians can then be exported into into NVIDIA Isaac Sim based on 3DGUT to simulate humanoid robots in generated environments.

Sherwin Bahmani1 year ago

Joint work with @TianchangS @jiawei6_ren @huangjh_hjh Yifeng Jiang @Haithem_Turki @taiyasaki @davelindell @ZGojcic @FidlerSanja @HuanLing6 @JunGao33210520 @xuanchi13

Sherwin Bahmani1 year ago

@jiawei6_ren @huangjh_hjh @Haithem_Turki @taiyasaki @DaveLindell @ZGojcic @FidlerSanja @HuanLing6 @JunGao33210520 @xuanchi13 Check out this very related work Bolt3D ( at ICCV next month by @StanSzymanowicz @jasonyzhang2 @_pratul_ @RuiqiGao @ArthurBrussee @holynski_ @rmbrualla @jon_barron @philipphenzler

Sherwin Bahmani1 year ago

@jiawei6_ren @huangjh_hjh @Haithem_Turki @taiyasaki @DaveLindell @ZGojcic @FidlerSanja @HuanLing6 @JunGao33210520 @xuanchi13 @StanSzymanowicz @jasonyzhang2 @_pratul_ @RuiqiGao @holynski_ @rmbrualla @jon_barron @philipphenzler Special thanks to @robrombach @andi_blatt and the rest of the team at @bfl_ml for open-sourcing FLUX ( We use FLUX to generate high-quality input images and lift them into 3D.

Matthias Niessner1 year ago

Amazing work!

Sherwin Bahmani1 year ago

Thanks Matthias 🍻

Kai He1 year ago

Strong 👍👍🥳

Sherwin Bahmani1 year ago

Not as strong as you 😌

Eduard Zamfir1 year ago

Uff 😮‍💨 sky is the limit! 🚀

Sherwin Bahmani1 year ago

Danke du Spruecheklopfer 🤪

Oliver Hahn1 year ago

Machine! Congrats 🚀

Sherwin Bahmani1 year ago

Danke Bro 😎

Lily Goli1 year ago

looks amazing!! 🎉

Sherwin Bahmani1 year ago

Thanks Lily 😊

Stephen J. Caggiano1 year ago

🔥❤️👏🏼

Ziya Erkoç1 year ago

Congrats, Sherwin!🎉🎉

Sherwin Bahmani1 year ago

Thanks Ziya 🫡

Tuan Ngo1 year ago

Amazing!!! Congrats Sherwin

Sherwin Bahmani1 year ago

Thanks big guy 🙏

Ison Ward1 year ago

Cool！That looks like a good time

Sai Tedla1 year ago

This is insane.

Phil Franco1 year ago

that looks cool!

Yuyang Zhao1 year ago

Amazing work! Congrats, Sherwin!

Sherwin Bahmani1 year ago

Thanks Yuyang 🤪

Hadi Alzayer1 year ago

Very cool work! Congrats on the release

Sherwin Bahmani1 year ago

Thanks Hadi 😊

:31 year ago

This is over 20 years ago:

Felix Taubner1 year ago

The 4D 🐐 🤩

Sherwin Bahmani1 year ago

No, you are the 4D 🐐😮‍💨

Mohamad Shahbazi1 year ago

Really cool work and amazing results Sherwin! :) Was wondering, how sensitive the 3DGS decoder is to the multivew inconsistencies?

Sherwin Bahmani1 year ago

Thanks! We learn to smoothen inconsistencies with LPIPS loss. Obviously, a more consistent diffusion model will make it easier to get consistent output. The current viewpoint change is mainly limited because the diffusion output gets more and more inconsistent far away from the input view. Hope to improve that!

Ethan Weber1 year ago

Congrats! I really like the ideas here. 🙂

Sherwin Bahmani1 year ago

Thanks Ethan 😊

Andrew Guo7 months ago

wow! this is cool

Related Videos

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.

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

294,519 views • 3 years ago

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

Alexandre Morgand

15,192 views • 8 months ago

📢📢GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction📢📢 Reconstructing high-fidelity 3D scenes from sparse RGB input is hard. It needs a strong 3D prior! We reformulate multi-view scene reconstruction as conditional 3D generation over overlapping spatial chunks, lifting posed image features into a generative shape prior via 3D conditioning. As an example prior, we build on Trellis2, and train it such that its reconstruction is pixel aligned and matches from all views. GenRecon achieves unprecedented reconstruction quality from any sparse RGB input sequence, even from a phone capture. The reconstruction also includes PBR materials which facilitates relighting and virtual object insertion. Amazing work by Katharina Schmid, Nicolas von Lützow, Jozef, Angela Dai

📢📢GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction📢📢 Reconstructing high-fidelity 3D scenes from sparse RGB input is hard. It needs a strong 3D prior! We reformulate multi-view scene reconstruction as conditional 3D generation over overlapping spatial chunks, lifting posed image features into a generative shape prior via 3D conditioning. As an example prior, we build on Trellis2, and train it such that its reconstruction is pixel aligned and matches from all views. GenRecon achieves unprecedented reconstruction quality from any sparse RGB input sequence, even from a phone capture. The reconstruction also includes PBR materials which facilitates relighting and virtual object insertion. Amazing work by Katharina Schmid, Nicolas von Lützow, Jozef, Angela Dai

Matthias Niessner

18,420 views • 4 months ago

Current 3D generative models are slow and low quality. We present GRM, a large-scale model that reconstructs 3D Gaussians in 0.1s and generates high-quality 3D assets from text or single images in a few seconds. Demo: 1/4

Current 3D generative models are slow and low quality. We present GRM, a large-scale model that reconstructs 3D Gaussians in 0.1s and generates high-quality 3D assets from text or single images in a few seconds. Demo: 1/4

Gordon Wetzstein

19,241 views • 2 years ago

Spatio-Temporal Reconstruction Model for Large-Scale Outdoor Scenes Contributions: • We propose STORM, the first feed-forward, self-supervised method for fast and accurate reconstruction of dynamic 3D scenes from sparse, multi-timestep, posed camera images. • Our bottom-up framework aggregates and transforms per-frame 3D Gaussian Splats into a cohesive scene representation, enabling self-supervised motion estimation. Furthermore, we introduce motion tokens that capture common motion primitives and regularize motion predictions, facilitating dynamic motion group segmentation without explicit motion or correspondence supervision. • We present several enhancements for in-the-wild scenarios, including sky modeling, camera exposure inconsistency handling, large novel-view extrapolation, and fine-grained human motions reconstruction, making STORM well-suited for real-world applications.

Spatio-Temporal Reconstruction Model for Large-Scale Outdoor Scenes Contributions: • We propose STORM, the first feed-forward, self-supervised method for fast and accurate reconstruction of dynamic 3D scenes from sparse, multi-timestep, posed camera images. • Our bottom-up framework aggregates and transforms per-frame 3D Gaussian Splats into a cohesive scene representation, enabling self-supervised motion estimation. Furthermore, we introduce motion tokens that capture common motion primitives and regularize motion predictions, facilitating dynamic motion group segmentation without explicit motion or correspondence supervision. • We present several enhancements for in-the-wild scenarios, including sky modeling, camera exposure inconsistency handling, large novel-view extrapolation, and fine-grained human motions reconstruction, making STORM well-suited for real-world applications.

MrNeRF

53,292 views • 1 year ago

V3D Video Diffusion Models are Effective 3D Generators Automatic 3D generation has recently attracted widespread attention. Recent methods have greatly accelerated the generation speed, but usually produce less-detailed objects due to limited model capacity or 3D data. Motivated by recent advancements in video diffusion models, we introduce V3D, which leverages the world simulation capacity of pre-trained video diffusion models to facilitate 3D generation. To fully unleash the potential of video diffusion to perceive the 3D world, we further introduce geometrical consistency prior and extend the video diffusion model to a multi-view consistent 3D generator. Benefiting from this, the state-of-the-art video diffusion model could be fine-tuned to generate 360degree orbit frames surrounding an object given a single image. With our tailored reconstruction pipelines, we can generate high-quality meshes or 3D Gaussians within 3 minutes. Furthermore, our method can be extended to scene-level novel view synthesis, achieving precise control over the camera path with sparse input views. Extensive experiments demonstrate the superior performance of the proposed approach, especially in terms of generation quality and multi-view consistency

V3D Video Diffusion Models are Effective 3D Generators Automatic 3D generation has recently attracted widespread attention. Recent methods have greatly accelerated the generation speed, but usually produce less-detailed objects due to limited model capacity or 3D data. Motivated by recent advancements in video diffusion models, we introduce V3D, which leverages the world simulation capacity of pre-trained video diffusion models to facilitate 3D generation. To fully unleash the potential of video diffusion to perceive the 3D world, we further introduce geometrical consistency prior and extend the video diffusion model to a multi-view consistent 3D generator. Benefiting from this, the state-of-the-art video diffusion model could be fine-tuned to generate 360degree orbit frames surrounding an object given a single image. With our tailored reconstruction pipelines, we can generate high-quality meshes or 3D Gaussians within 3 minutes. Furthermore, our method can be extended to scene-level novel view synthesis, achieving precise control over the camera path with sparse input views. Extensive experiments demonstrate the superior performance of the proposed approach, especially in terms of generation quality and multi-view consistency

AK

31,997 views • 2 years ago

DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior paper page: present DreamCraft3D, a hierarchical 3D content generation method that produces high-fidelity and coherent 3D objects. We tackle the problem by leveraging a 2D reference image to guide the stages of geometry sculpting and texture boosting. A central focus of this work is to address the consistency issue that existing works encounter. To sculpt geometries that render coherently, we perform score distillation sampling via a view-dependent diffusion model. This 3D prior, alongside several training strategies, prioritizes the geometry consistency but compromises the texture fidelity. We further propose Bootstrapped Score Distillation to specifically boost the texture. We train a personalized diffusion model, Dreambooth, on the augmented renderings of the scene, imbuing it with 3D knowledge of the scene being optimized. The score distillation from this 3D-aware diffusion prior provides view-consistent guidance for the scene. Notably, through an alternating optimization of the diffusion prior and 3D scene representation, we achieve mutually reinforcing improvements: the optimized 3D scene aids in training the scene-specific diffusion model, which offers increasingly view-consistent guidance for 3D optimization. The optimization is thus bootstrapped and leads to substantial texture boosting. With tailored 3D priors throughout the hierarchical generation, DreamCraft3D generates coherent 3D objects with photorealistic renderings, advancing the state-of-the-art in 3D content generation.

DreamCraft3D: Hierarchical 3D Generation with Bootstrapped Diffusion Prior paper page: present DreamCraft3D, a hierarchical 3D content generation method that produces high-fidelity and coherent 3D objects. We tackle the problem by leveraging a 2D reference image to guide the stages of geometry sculpting and texture boosting. A central focus of this work is to address the consistency issue that existing works encounter. To sculpt geometries that render coherently, we perform score distillation sampling via a view-dependent diffusion model. This 3D prior, alongside several training strategies, prioritizes the geometry consistency but compromises the texture fidelity. We further propose Bootstrapped Score Distillation to specifically boost the texture. We train a personalized diffusion model, Dreambooth, on the augmented renderings of the scene, imbuing it with 3D knowledge of the scene being optimized. The score distillation from this 3D-aware diffusion prior provides view-consistent guidance for the scene. Notably, through an alternating optimization of the diffusion prior and 3D scene representation, we achieve mutually reinforcing improvements: the optimized 3D scene aids in training the scene-specific diffusion model, which offers increasingly view-consistent guidance for 3D optimization. The optimization is thus bootstrapped and leads to substantial texture boosting. With tailored 3D priors throughout the hierarchical generation, DreamCraft3D generates coherent 3D objects with photorealistic renderings, advancing the state-of-the-art in 3D content generation.

AK

161,530 views • 3 years ago

3D scene reconstructions by NVIDIA. ArtiFixer - repairs artifacts and extends sparse views via Wan 2.1. - high-fidelity inpainting in occluded regions - gens hundreds of consistent frames in a single pass - 3D Gaussian Splatting for navigable scene reconstruction Makes the 3D environment look photorealistic and fully navigable for VR/AR. It basically turns a broken 3D model into a polished, professional scene.

3D scene reconstructions by NVIDIA. ArtiFixer - repairs artifacts and extends sparse views via Wan 2.1. - high-fidelity inpainting in occluded regions - gens hundreds of consistent frames in a single pass - 3D Gaussian Splatting for navigable scene reconstruction Makes the 3D environment look photorealistic and fully navigable for VR/AR. It basically turns a broken 3D model into a polished, professional scene.

Wildminder

16,086 views • 4 months ago

🚀Turn Single Image into 3D Human🚀 #GeneMAN is a generalizable single-image 3D human reconstruction framework that turns in-the-wild images into high-quality 3D humans with ease 🔗Project: 📜Paper: 🧑‍💻Code:

🚀Turn Single Image into 3D Human🚀 #GeneMAN is a generalizable single-image 3D human reconstruction framework that turns in-the-wild images into high-quality 3D humans with ease 🔗Project: 📜Paper: 🧑‍💻Code:

Ziwei Liu

26,953 views • 1 year ago

Synthesizing worlds with video diffusion models is often inconsistent — moving the camera back and forth leads to different scenes. We propose 🌐𝗪𝗼𝗿𝗹𝗱𝗠𝗲𝗺, a memory-based approach that ensures consistent world simulation without relying on explicit 3D reconstruction.

Synthesizing worlds with video diffusion models is often inconsistent — moving the camera back and forth leads to different scenes. We propose 🌐𝗪𝗼𝗿𝗹𝗱𝗠𝗲𝗺, a memory-based approach that ensures consistent world simulation without relying on explicit 3D reconstruction.

Xingang Pan

19,413 views • 1 year ago

📢📢 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 views • 1 year ago

Introducing Code Video Model 🎬 Amazing work led by Ziyu Wan and the team! Agentic coding as the engine for controllable video generation. Scene layout, motion, timing, camera trajectories, and physical dynamics, all in editable code. A coding agent turns prompts into executable programs that render proxy videos. Our training-free early-step conditioning lets a pretrained video model follow the proxy’s structure while adopting a reference image’s appearance. From physics-grounded motion and bullet time to 3D/4D reconstruction and robotics simulation, it all starts with code. Watch the demo and explore more visual results 👇 Project page: Paper: Code: Data: Youtube:

Introducing Code Video Model 🎬 Amazing work led by Ziyu Wan and the team! Agentic coding as the engine for controllable video generation. Scene layout, motion, timing, camera trajectories, and physical dynamics, all in editable code. A coding agent turns prompts into executable programs that render proxy videos. Our training-free early-step conditioning lets a pretrained video model follow the proxy’s structure while adopting a reference image’s appearance. From physics-grounded motion and bullet time to 3D/4D reconstruction and robotics simulation, it all starts with code. Watch the demo and explore more visual results 👇 Project page: Paper: Code: Data: Youtube:

Gordon (Guocheng) Qian

39,496 views • 7 days ago

SAM 3D enables accurate 3D reconstruction from a single image, supporting real-world applications in editing, robotics, and interactive scene generation. Matt, a SAM 3D researcher, explains how the two-model design makes this possible for both people and complex environments. 🔗 Read the SAM 3D Objects research paper: 🔗 Read the SAM 3D Body research paper:

SAM 3D enables accurate 3D reconstruction from a single image, supporting real-world applications in editing, robotics, and interactive scene generation. Matt, a SAM 3D researcher, explains how the two-model design makes this possible for both people and complex environments. 🔗 Read the SAM 3D Objects research paper: 🔗 Read the SAM 3D Body research paper:

AI at Meta

18,031 views • 11 months ago

Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors paper page: Most 3D generation research focuses on up-projecting 2D foundation models into the 3D space, either by minimizing 2D Score Distillation Sampling (SDS) loss or fine-tuning on multi-view datasets. Without explicit 3D priors, these methods often lead to geometric anomalies and multi-view inconsistency. Recently, researchers have attempted to improve the genuineness of 3D objects by directly training on 3D datasets, albeit at the cost of low-quality texture generation due to the limited texture diversity in 3D datasets. To harness the advantages of both approaches, we propose Bidirectional Diffusion(BiDiff), a unified framework that incorporates both a 3D and a 2D diffusion process, to preserve both 3D fidelity and 2D texture richness, respectively. Moreover, as a simple combination may yield inconsistent generation results, we further bridge them with novel bidirectional guidance. In addition, our method can be used as an initialization of optimization-based models to further improve the quality of 3D model and efficiency of optimization, reducing the generation process from 3.4 hours to 20 minutes. Experimental results have shown that our model achieves high-quality, diverse, and scalable 3D generation.

Text-to-3D Generation with Bidirectional Diffusion using both 2D and 3D priors paper page: Most 3D generation research focuses on up-projecting 2D foundation models into the 3D space, either by minimizing 2D Score Distillation Sampling (SDS) loss or fine-tuning on multi-view datasets. Without explicit 3D priors, these methods often lead to geometric anomalies and multi-view inconsistency. Recently, researchers have attempted to improve the genuineness of 3D objects by directly training on 3D datasets, albeit at the cost of low-quality texture generation due to the limited texture diversity in 3D datasets. To harness the advantages of both approaches, we propose Bidirectional Diffusion(BiDiff), a unified framework that incorporates both a 3D and a 2D diffusion process, to preserve both 3D fidelity and 2D texture richness, respectively. Moreover, as a simple combination may yield inconsistent generation results, we further bridge them with novel bidirectional guidance. In addition, our method can be used as an initialization of optimization-based models to further improve the quality of 3D model and efficiency of optimization, reducing the generation process from 3.4 hours to 20 minutes. Experimental results have shown that our model achieves high-quality, diverse, and scalable 3D generation.

AK

38,853 views • 2 years ago

(1/N) Will this be the BERT/GPT moment for 3D vision？ Finally, unsupervised pre-training for 3D works. Led by Qitao Zhao , we present E-RayZer — a fully self-supervised 3D reconstruction model that: 🔥Matches or surpasses supervised methods like VGGT 👀Learns transferable 3D representations, outperforming CroCo, VideoMAE, and DINO 📈Scales with more unlabeled data A new recipe for scalable 3D foundation models.

(1/N) Will this be the BERT/GPT moment for 3D vision？ Finally, unsupervised pre-training for 3D works. Led by Qitao Zhao , we present E-RayZer — a fully self-supervised 3D reconstruction model that: 🔥Matches or surpasses supervised methods like VGGT 👀Learns transferable 3D representations, outperforming CroCo, VideoMAE, and DINO 📈Scales with more unlabeled data A new recipe for scalable 3D foundation models.

Hanwen Jiang

58,737 views • 10 months ago

We’re open-sourcing HY-World 2.0, a multimodal world model that generates, reconstructs, and simulates interactive *3D worlds* from text, images, and videos. Outputs can be integrated into game engines and embodied simulation pipelines. Key highlights: 🔹 One-click world generation Turn text or image into interactive 3D worlds automatically. 🔹 Pipeline-ready 3D outputs Editable 3D worlds for Unity and Unreal Engine, with standard 3D exports including mesh, 3DGS, and point clouds. 🔹 Unified world model system One model family for world generation and reconstruction across synthetic and real-world scenes. 🔹 Interactive character mode Explore generated 3D worlds in real time with physics-aware movement and collision support. ✨ Apply for access: 🔗 GitHub: 🤗 Hugging Face: 📄 Technical Report:

We’re open-sourcing HY-World 2.0, a multimodal world model that generates, reconstructs, and simulates interactive 3D worlds from text, images, and videos. Outputs can be integrated into game engines and embodied simulation pipelines. Key highlights: 🔹 One-click world generation Turn text or image into interactive 3D worlds automatically. 🔹 Pipeline-ready 3D outputs Editable 3D worlds for Unity and Unreal Engine, with standard 3D exports including mesh, 3DGS, and point clouds. 🔹 Unified world model system One model family for world generation and reconstruction across synthetic and real-world scenes. 🔹 Interactive character mode Explore generated 3D worlds in real time with physics-aware movement and collision support. ✨ Apply for access: 🔗 GitHub: 🤗 Hugging Face: 📄 Technical Report:

Tencent Hy

373,023 views • 6 months ago

HunyuanWorld-Voyager is here and fully open-source! The world’s first ultra-long-range world model with native 3D reconstruction, redefining AI-driven spatial intelligence for VR, gaming, and simulations. ✅Direct 3D Output: Exports point cloud videos to 3D formats without tools like COLMAP, enabling instant 3D application use. ✅Innovative 3D Memory: Introduces a scalable world caching mechanism, ensuring geometric consistency across any camera trajectory. ✅Top-Ranked Performance: #1 on Stanford’s WorldScore, excelling in video generation and 3D reconstruction benchmarks.( Built on HunyuanWorld 1.0, Voyager blends video generation with 3D modeling, delivering camera-controlled, high-fidelity RGB-D sequences. Control scenes via keyboard or joystick for unmatched 3D consistency. Explore now: 🌐Project Page: 🔗GitHub: 🤗HuggingFace: 📝Technical Details:

HunyuanWorld-Voyager is here and fully open-source! The world’s first ultra-long-range world model with native 3D reconstruction, redefining AI-driven spatial intelligence for VR, gaming, and simulations. ✅Direct 3D Output: Exports point cloud videos to 3D formats without tools like COLMAP, enabling instant 3D application use. ✅Innovative 3D Memory: Introduces a scalable world caching mechanism, ensuring geometric consistency across any camera trajectory. ✅Top-Ranked Performance: #1 on Stanford’s WorldScore, excelling in video generation and 3D reconstruction benchmarks.( Built on HunyuanWorld 1.0, Voyager blends video generation with 3D modeling, delivering camera-controlled, high-fidelity RGB-D sequences. Control scenes via keyboard or joystick for unmatched 3D consistency. Explore now: 🌐Project Page: 🔗GitHub: 🤗HuggingFace: 📝Technical Details:

Tencent Hy

198,391 views • 1 year ago

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

LLM-grounded Diffusion: Enhancing Prompt Understanding of Text-to-Image Diffusion Models with Large Language Models paper page: github: Recent advancements in text-to-image generation with diffusion models have yielded remarkable results synthesizing highly realistic and diverse images. However, these models still encounter difficulties when generating images from prompts that demand spatial or common sense reasoning. We propose to equip diffusion models with enhanced reasoning capabilities by using off-the-shelf pretrained large language models (LLMs) in a novel two-stage generation process. First, we adapt an LLM to be a text-guided layout generator through in-context learning. When provided with an image prompt, an LLM outputs a scene layout in the form of bounding boxes along with corresponding individual descriptions. Second, we steer a diffusion model with a novel controller to generate images conditioned on the layout. Both stages utilize frozen pretrained models without any LLM or diffusion model parameter optimization. We validate the superiority of our design by demonstrating its ability to outperform the base diffusion model in accurately generating images according to prompts that necessitate both language and spatial reasoning. Additionally, our method naturally allows dialog-based scene specification and is able to handle prompts in a language that is not well-supported by the underlying diffusion model.

AK

83,681 views • 3 years ago

🚀 Our NeurIPS '24 work, Large Spatial Model (LSM), is here! LSM performs semantic 3D reconstruction in just 0.1s, processing unposed data via feed-forward 3D reconstruction. 👉It leverages large-scale 3D datasets with minimal annotations, defining a 3D latent space. We are continuously exploring how this explicit 3D representation can further enhance reasoning and robotic learning. 🔗 Try our online Gradio demo with your own data at #NeurIPS2024 #3DReconstruction

🚀 Our NeurIPS '24 work, Large Spatial Model (LSM), is here! LSM performs semantic 3D reconstruction in just 0.1s, processing unposed data via feed-forward 3D reconstruction. 👉It leverages large-scale 3D datasets with minimal annotations, defining a 3D latent space. We are continuously exploring how this explicit 3D representation can further enhance reasoning and robotic learning. 🔗 Try our online Gradio demo with your own data at #NeurIPS2024 #3DReconstruction

Zhiwen(Aaron) Fan

44,051 views • 2 years ago