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.

Introducing SAM 3D, the newest addition to the SAM collection, bringing common sense 3D understanding of everyday images. SAM 3D includes two models: 🛋️ SAM 3D Objects for object and scene reconstruction 🧑‍🤝‍🧑 SAM 3D Body for human pose and shape estimation Both models achieve state-of-the-art performance transforming static... show more

AI at Meta

864,096 subscribers

861,526 views • 10 months ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

35 Comments

AI at Meta10 months ago

We’re sharing model checkpoints, an evaluation benchmark, human body training data, and inference code with the community to support creative applications in fields like robotics, interactive media, science, sports medicine, and beyond. 🔗 SAM 3D Body: 🔗 SAM 3D Objects:

AI at Meta10 months 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:

Alex Yankov10 months ago

Pose estimation works even for illustrations. Completely unreal.

Priyanshu10 months ago

This is what you were cooking? Damnn

Hari10 months ago

Now is the best time to be a developer.. design any 3D games, design any retro games you would like to and obviously anything you can imagine make it to the world with the help of prompt and introductions of models like SAM 3D 📝

Subhankar10 months ago

The quiet genius of SAM 3D: it doesn’t hallucinate geometry, it infers it from the same way humans do—with everyday priors baked in. Releasing both object and body models together means mixed-scene understanding just became democratized. This is foundation-model embodiment done right.

Rabu10 months ago

This is huge 🚀 SAM 3D adds a whole new dimension to the SAM lineup literally. Turning everyday 2D images into detailed 3D reconstructions is a game changer for creators developers and immersive experiences. SAM 3D Objects plus SAM 3D Body delivering state of the art reconstruction and pose understanding opens the door to next level AR design and content creation. Excited to see where this goes.

gabriel10 months ago

this is so cool!

NOKOL10 months ago

Wow ok this is sick. Greetings to mark for this accomplishment of his crew. Will open some business opportunities 👀☕

ILYA10 months ago

Who wants to work with me on smth using this?

logix10 months ago

i sincerely thank you for your OSS stance

Ron Adin10 months ago

ok..

Shuming Hu10 months ago

very cool! congrats to the team!

Zhengzhong Tu10 months ago

Long life FAIR

SKYLINE🥷10 months ago

3d reconstructions gonna make edits way cooler

BrennstoffBernd10 months ago

That is quite impressive! Nice work. I can imagine that some function like the meta quest hyperspace can also benefit from models like these.

YUHONG LI10 months ago

Is there a website for me to play around

webXOS Software10 months ago

3I ATLAS TESTING:

Nikol.pls❤️‍🔥⚡️10 months ago

Yes! Needed for a current project

hk_next10 months ago

Gaussian Splatting changed the game for digital twins, but it couldn’t separate objects. SAM 3D plugs that last hole—and real-world robotics is about to speed up fast.

Pourush Siddharth10 months ago

RIP the 'Junior 3D Modeler' role. SAM 3D turns 2D photos into textured meshes instantly. The asset pipeline just went from 100 hours to 100 milliseconds. Barrier to entry: -90%. The era of the 'One-Person AAA Studio' is here.

Vincent Cho10 months ago

Very impressive, congrats SAM 3D team!

max iuliani ハリ🌐10 months ago

Impressive work with SAM 3D — segmentation and reconstruction are essential steps. But the next frontier isn’t understanding objects. It’s understanding *humans* — context, emotion, intent, risk patterns, and situational meaning. I’m working on a human-aligned architecture (HARI) focused on predictive empathy, contextual harmonization and safe coexistence between humans and AI. When your models will need to interpret *real life* instead of just 3D scenes, let me know. We may have something you’ll want to see. #AISafety #HumanAI #Alignment #AIforGood

Mat Komeng10 months ago

Tested. nothing to see.

braeden hall10 months ago

@ionthedev could maybe get drone model if we still need

Chris DeWeese10 months ago

@ayaboch

Quentin Merabet10 months ago

What the hell

Mr Ait10 months ago

Cool

岡田泰彦｜光邦｜AI×印刷エヴァンジェリスト10 months ago

This is a huge leap for visual understanding—turning everyday photos into accurate 3D reconstructions feels like sci-fi becoming real. 日常の写真が正確な3Dに変わるなんて、本当にSFが現実になった瞬間ですね。

competitivethinker10 months ago

If this really works, it is a much bigger deal than it seems

Maulik Maanche @10 months ago

I wish to use it in blender. Great times

Shubham Verma10 months ago

Meet SAM 3D, the latest in the SAM family, bringing common-sense 3D understanding to everyday images. 🛋️ Objects for scene reconstruction 🧑‍🤝‍🧑 Body for human pose and shape Both set a new state of the art in transforming 2D images into vivid, accurate 3D.

Benjamin Couture10 months ago

@felixdigit

Shubham Ranga10 months ago

super cool, 2d pics now spinning full-blown reality engines, unreal upgrade.

Arsi10 months ago

3d bodies from flat pics still blows my mind

Related Videos

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 • 10 months ago

Today we’re excited to unveil a new generation of Segment Anything Models: 1️⃣ SAM 3 enables detecting, segmenting and tracking of objects across images and videos, now with short text phrases and exemplar prompts. 🔗 Learn more about SAM 3: 2️⃣ SAM 3D brings the model collection into the 3rd dimension to enable precise reconstruction of 3D objects and people from a single 2D image. 🔗 Learn more about SAM 3D: These models offer innovative capabilities and unique tools for developers and researchers to create, experiment and uplevel media workflows.

Today we’re excited to unveil a new generation of Segment Anything Models: 1️⃣ SAM 3 enables detecting, segmenting and tracking of objects across images and videos, now with short text phrases and exemplar prompts. 🔗 Learn more about SAM 3: 2️⃣ SAM 3D brings the model collection into the 3rd dimension to enable precise reconstruction of 3D objects and people from a single 2D image. 🔗 Learn more about SAM 3D: These models offer innovative capabilities and unique tools for developers and researchers to create, experiment and uplevel media workflows.

AI at Meta

1,091,560 views • 10 months ago

SAM 3D is helping advance the future of rehabilitation. See how researchers at Carnegie Mellon University are using SAM 3D to capture and analyze human movement in clinical settings, opening the doors to personalized, data-driven insights in the recovery process. 🔗 Learn more about SAM 3D:

SAM 3D is helping advance the future of rehabilitation. See how researchers at Carnegie Mellon University are using SAM 3D to capture and analyze human movement in clinical settings, opening the doors to personalized, data-driven insights in the recovery process. 🔗 Learn more about SAM 3D:

AI at Meta

69,058 views • 10 months ago

Just made Pascal's item catalog infinite. Pulling 3D objects straight out of images, live in the browser. SAM 3D Objects + Three.js + WebGPU. 👇

Just made Pascal's item catalog infinite. Pulling 3D objects straight out of images, live in the browser. SAM 3D Objects + Three.js + WebGPU. 👇

Wawa Sensei

46,888 views • 5 months ago

LLaVA-3D A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness Recent advancements in Large Multimodal Models (LMMs) have greatly enhanced their proficiency in 2D visual understanding tasks, enabling them to effectively process and understand images and videos. However, the development of LMMs with 3D-awareness for 3D scene understanding has been hindered by the lack of large-scale 3D vision-language datasets and powerful 3D encoders. In this paper, we introduce a simple yet effective framework called LLaVA-3D. Leveraging the strong 2D understanding priors from LLaVA, our LLaVA-3D efficiently adapts LLaVA for 3D scene understanding without compromising 2D understanding capabilities. To achieve this, we employ a simple yet effective representation, 3D Patch, which connects 2D CLIP patch features with their corresponding positions in 3D space. By integrating the 3D Patches into 2D LMMs and employing joint 2D and 3D vision-language instruction tuning, we establish a unified architecture for both 2D image understanding and 3D scene understanding. Experimental results show that LLaVA-3D converges 3.5x faster than existing 3D LMMs when trained on 3D vision-language datasets. Moreover, LLaVA-3D not only achieves state-of-the-art performance across various 3D tasks but also maintains comparable 2D image understanding and vision-language conversation capabilities with LLaVA.

LLaVA-3D A Simple yet Effective Pathway to Empowering LMMs with 3D-awareness Recent advancements in Large Multimodal Models (LMMs) have greatly enhanced their proficiency in 2D visual understanding tasks, enabling them to effectively process and understand images and videos. However, the development of LMMs with 3D-awareness for 3D scene understanding has been hindered by the lack of large-scale 3D vision-language datasets and powerful 3D encoders. In this paper, we introduce a simple yet effective framework called LLaVA-3D. Leveraging the strong 2D understanding priors from LLaVA, our LLaVA-3D efficiently adapts LLaVA for 3D scene understanding without compromising 2D understanding capabilities. To achieve this, we employ a simple yet effective representation, 3D Patch, which connects 2D CLIP patch features with their corresponding positions in 3D space. By integrating the 3D Patches into 2D LMMs and employing joint 2D and 3D vision-language instruction tuning, we establish a unified architecture for both 2D image understanding and 3D scene understanding. Experimental results show that LLaVA-3D converges 3.5x faster than existing 3D LMMs when trained on 3D vision-language datasets. Moreover, LLaVA-3D not only achieves state-of-the-art performance across various 3D tasks but also maintains comparable 2D image understanding and vision-language conversation capabilities with LLaVA.

AK

41,747 views • 2 years ago

3D Congealing 3D-Aware Image Alignment in the Wild We propose 3D Congealing, a novel problem of 3D-aware alignment for 2D images capturing semantically similar objects. Given a collection of unlabeled Internet images, our goal is to associate the shared semantic parts

3D Congealing 3D-Aware Image Alignment in the Wild We propose 3D Congealing, a novel problem of 3D-aware alignment for 2D images capturing semantically similar objects. Given a collection of unlabeled Internet images, our goal is to associate the shared semantic parts

AK

16,325 views • 2 years ago

3D Trailer for Sam Raimi’s “SEND HELP”. This will be Raimi’s third 3D film following Oz the Great and Powerful & Doctor Strange in the Multiverse of Madness.

3D Trailer for Sam Raimi’s “SEND HELP”. This will be Raimi’s third 3D film following Oz the Great and Powerful & Doctor Strange in the Multiverse of Madness.

Sam Raimi Updates

41,014 views • 9 months ago

I am releasing the “MLSharp” app for macOS. MLSharp is a premium macOS application that transforms 2D images into immersive 3D Gaussian Splatting (3DGS) models using state-of-the-art AI.

I am releasing the “MLSharp” app for macOS. MLSharp is a premium macOS application that transforms 2D images into immersive 3D Gaussian Splatting (3DGS) models using state-of-the-art AI.

むっちゃん

91,080 views • 6 months ago

We are introducing Hunyuan3D-Part: an open-source part-level 3D shape generation model that outperforms all existing open and close-source models. Highlights: 🔹P3-SAM: The industry's first native 3D part segmentation model. 🔹X-Part: A part generation model that achieves state-of-the-art results in controllability and shape quality. Key-features: 1️⃣Eliminates the use of 2D SAM during training, relying solely on a large-scale dataset with 3.7 million shapes and clean part annotations. 2️⃣Introduces a new automated segmentation pipeline in 3D without user intervention. 3️⃣Implements a diffusion-based part decomposition pipeline utilizing both geometry and semantic clues. Code: Weights: Tech reports: 🔸P3-SAM： → Paper: → Project page: 🔸X-Part： → Paper: → Project page: Try it now： → (Light version) Hugging Face demo: → (Full version) Hunyuan3D Studio:

We are introducing Hunyuan3D-Part: an open-source part-level 3D shape generation model that outperforms all existing open and close-source models. Highlights: 🔹P3-SAM: The industry's first native 3D part segmentation model. 🔹X-Part: A part generation model that achieves state-of-the-art results in controllability and shape quality. Key-features: 1️⃣Eliminates the use of 2D SAM during training, relying solely on a large-scale dataset with 3.7 million shapes and clean part annotations. 2️⃣Introduces a new automated segmentation pipeline in 3D without user intervention. 3️⃣Implements a diffusion-based part decomposition pipeline utilizing both geometry and semantic clues. Code: Weights: Tech reports: 🔸P3-SAM： → Paper: → Project page: 🔸X-Part： → Paper: → Project page: Try it now： → (Light version) Hugging Face demo: → (Full version) Hunyuan3D Studio:

Tencent Hy

72,640 views • 1 year ago

An Object is Worth 64x64 Pixels: Generating 3D Object via Image Diffusion discuss: We introduce a new approach for generating realistic 3D models with UV maps through a representation termed "Object Images." This approach encapsulates surface geometry, appearance, and patch structures within a 64x64 pixel image, effectively converting complex 3D shapes into a more manageable 2D format. By doing so, we address the challenges of both geometric and semantic irregularity inherent in polygonal meshes. This method allows us to use image generation models, such as Diffusion Transformers, directly for 3D shape generation. Evaluated on the ABO dataset, our generated shapes with patch structures achieve point cloud FID comparable to recent 3D generative models, while naturally supporting PBR material generation.

An Object is Worth 64x64 Pixels: Generating 3D Object via Image Diffusion discuss: We introduce a new approach for generating realistic 3D models with UV maps through a representation termed "Object Images." This approach encapsulates surface geometry, appearance, and patch structures within a 64x64 pixel image, effectively converting complex 3D shapes into a more manageable 2D format. By doing so, we address the challenges of both geometric and semantic irregularity inherent in polygonal meshes. This method allows us to use image generation models, such as Diffusion Transformers, directly for 3D shape generation. Evaluated on the ABO dataset, our generated shapes with patch structures achieve point cloud FID comparable to recent 3D generative models, while naturally supporting PBR material generation.

AK

66,769 views • 2 years ago

Introducing HO-Cap: A Capture System and Dataset for 3D Reconstruction and Pose Tracking of Hand-Object Interaction. We built a multi-camera system and a semi-automatic method for annotating the shape and pose of hands and objects Project page:

Introducing HO-Cap: A Capture System and Dataset for 3D Reconstruction and Pose Tracking of Hand-Object Interaction. We built a multi-camera system and a semi-automatic method for annotating the shape and pose of hands and objects Project page:

Yu Xiang

57,561 views • 1 year ago

Multi-modal #LLMs understand a lot about humans. But do they understand our 3D pose? We train #PoseGPT to estimate, generate, and reason about 3D human pose (#SMPL) in images and text. This is the first true foundation model for understanding 3D humans.

Multi-modal #LLMs understand a lot about humans. But do they understand our 3D pose? We train #PoseGPT to estimate, generate, and reason about 3D human pose (#SMPL) in images and text. This is the first true foundation model for understanding 3D humans.

Michael Black

81,430 views • 2 years ago

Collecting a high quality dataset with 4M unique phrases and 52M corresponding object masks helped SAM 3 achieve 2x the performance of baseline models. Kate, a researcher on SAM 3, explains how the data engine made this leap possible. 🔗 Read the SAM 3 research paper:

Collecting a high quality dataset with 4M unique phrases and 52M corresponding object masks helped SAM 3 achieve 2x the performance of baseline models. Kate, a researcher on SAM 3, explains how the data engine made this leap possible. 🔗 Read the SAM 3 research paper:

AI at Meta

39,436 views • 10 months ago

Introducing Wonder 3D, a new generative AI model in Flow Studio for fast and detailed 3D asset creation. 💬 Text to 3D — Watch your words take shape. Turn a simple prompt into fully textured 3D characters and props. 📸 Image to 3D — Add depth to your visuals. Transform concept art or reference images into dimensional 3D models. 🖼️ Text to Image — Visualize ideas early. Turn descriptions into concept imagery before converting them into 3D. Built for endless storytelling. Give your imagination some dimension. Available in all tiers. Find it under Wonder Tools. 👉 Get started: 💡 Learn more:

Introducing Wonder 3D, a new generative AI model in Flow Studio for fast and detailed 3D asset creation. 💬 Text to 3D — Watch your words take shape. Turn a simple prompt into fully textured 3D characters and props. 📸 Image to 3D — Add depth to your visuals. Transform concept art or reference images into dimensional 3D models. 🖼️ Text to Image — Visualize ideas early. Turn descriptions into concept imagery before converting them into 3D. Built for endless storytelling. Give your imagination some dimension. Available in all tiers. Find it under Wonder Tools. 👉 Get started: 💡 Learn more:

Autodesk Flow Studio

107,916 views • 7 months ago

oh my.. hows this even possible Meta just dropped SAM 3D, you can auto select any object in still image and.. turn them into high quality 3D model the model is free to download and test in playground link in comments

oh my.. hows this even possible Meta just dropped SAM 3D, you can auto select any object in still image and.. turn them into high quality 3D model the model is free to download and test in playground link in comments

el.cine

795,982 views • 10 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

Nano Banana 2, new state of the art in image generation and editing combined with Gemini’s real-world knowledge! You can simulate 3D CAD models purely through images. From sketch to real object!

Nano Banana 2, new state of the art in image generation and editing combined with Gemini’s real-world knowledge! You can simulate 3D CAD models purely through images. From sketch to real object!

koray kavukcuoglu

87,809 views • 7 months ago

WHAM defines the new state of the art in 3D human pose estimation from video. By a large margin. It’s fast, accurate, and it computes human pose in world coordinates. It’s also the first video-based method to be more accurate than single-image methods. 1/8

WHAM defines the new state of the art in 3D human pose estimation from video. By a large margin. It’s fast, accurate, and it computes human pose in world coordinates. It’s also the first video-based method to be more accurate than single-image methods. 1/8

Michael Black

118,503 views • 2 years ago

Introducing the TapNow x Marble WORLD Model! Transform static images into explorable environments with infinite angularity. The 3D redraw feature allows you to reimagine and transform objects within the generated space. Click the link below for more details.

Introducing the TapNow x Marble WORLD Model! Transform static images into explorable environments with infinite angularity. The 3D redraw feature allows you to reimagine and transform objects within the generated space. Click the link below for more details.

TapNow

2,652,996 views • 5 months 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