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。

Our SOTA world model, Echo, reconstructs high-quality scenes from sparse inputs and fills missing areas using its world knowledge. That means far less capture, data, and prep to produce geometrically consistent, multi-room reconstructions. We can’t wait to share it. Public release coming soon

SpAItial AI

4,738 subscribers

10,999 次观看 • 3 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

9 条评论

AI Mastery Guide 的头像

AI Mastery Guide2 个月前

Filling in missing areas with world knowledge instead of raw capture could cut scanning time down a lot.

Agoran 的头像

Agoran2 个月前

Can you get architecs as clients so they can mass produce large nice sturdy properties for low low prices?

SpAItial AI 的头像

SpAItial AI3 个月前

Try Echo today: App: Developer Api: Create with agents:

Ella Tech & Tool 的头像

Ella Tech & Tool2 个月前

This is wild Less data, better results. Can’t wait for Echo

Victoria Scott $ 的头像

Victoria Scott $2 个月前

"Echo-2 looks incredible. Moving from flat video generations to physically grounded, interactive 3D spaces is a massive leap forward."

Bblack 的头像

Bblack2 个月前

Less capture, more world. Echo is about to make 3D effortless 🔥

Alice The Ai Expert 的头像

Alice The Ai Expert2 个月前

Amazing Less capture, more world. Echo turns sparse inputs into full, geometrically consistent scenes.

RAZA | AI EXPLORER 的头像

RAZA | AI EXPLORER2 个月前

World models are unlocking incredible possibilities. Excited to see Echo in action.

willie 的头像

willie2 个月前

This is insane. Can you share the details of 'Sparse inputs' - how many panoramas to create the scene in the demo?

相关视频

Large, consistent, and high-quality 3D worlds from simple iPhone recordings. Works for multi-room scenes, fills missing gaps consistently while maintaining high quality, and it’s super easy to use. Now in beta with partners and coming soon to the SpAItial API. Link to the scenes in the comments so you can navigate this world yourself 👇🏽

Large, consistent, and high-quality 3D worlds from simple iPhone recordings. Works for multi-room scenes, fills missing gaps consistently while maintaining high quality, and it’s super easy to use. Now in beta with partners and coming soon to the SpAItial API. Link to the scenes in the comments so you can navigate this world yourself 👇🏽

SpAItial AI

78,785 次观看 • 1 个月前

NEWS: Waymo has introduced the Waymo World Model, "a frontier generative model built on Google DeepMind’s Genie 3 that sets a new bar for large-scale, hyper-realistic autonomous driving simulation." "By simulating the “impossible”, we proactively prepare the Waymo Driver for some of the most rare and complex scenarios—from tornadoes to planes landing on freeways—long before it encounters them in the real world. The model’s architecture offers high controllability, allowing our engineers to modify simulations with simple language prompts, driving inputs, and scene layouts. Notably, the Waymo World Model generates high-fidelity, multi-sensor outputs that include both camera and lidar data. This combination of broad world knowledge, fine-grained controllability, and multi-modal realism enhances Waymo’s ability to safely scale our service across more places and new driving environments. In the following sections we showcase the Waymo World Model in action, featuring simulations of the Waymo Driver navigating diverse rare edge-case scenarios." More info below.

NEWS: Waymo has introduced the Waymo World Model, "a frontier generative model built on Google DeepMind’s Genie 3 that sets a new bar for large-scale, hyper-realistic autonomous driving simulation." "By simulating the “impossible”, we proactively prepare the Waymo Driver for some of the most rare and complex scenarios—from tornadoes to planes landing on freeways—long before it encounters them in the real world. The model’s architecture offers high controllability, allowing our engineers to modify simulations with simple language prompts, driving inputs, and scene layouts. Notably, the Waymo World Model generates high-fidelity, multi-sensor outputs that include both camera and lidar data. This combination of broad world knowledge, fine-grained controllability, and multi-modal realism enhances Waymo’s ability to safely scale our service across more places and new driving environments. In the following sections we showcase the Waymo World Model in action, featuring simulations of the Waymo Driver navigating diverse rare edge-case scenarios." More info below.

Sawyer Merritt

133,382 次观看 • 8 个月前

📸 A few photos. ⚡ A few seconds. A high-quality 3D scene！🎬 Meet Topos-Lite, our new 3D Gaussian Splatting model. With just a few sparse multi-view images, it can generate a high-quality, renderable 3D scene in few seconds We’re excited to share more soon. The technical report is coming shortly — stay tuned and stay in touch. 👀 #ToposLite #3DGS #3DGeneration #SpatialAI #AIGC #photograph #3DArtwork

📸 A few photos. ⚡ A few seconds. A high-quality 3D scene！🎬 Meet Topos-Lite, our new 3D Gaussian Splatting model. With just a few sparse multi-view images, it can generate a high-quality, renderable 3D scene in few seconds We’re excited to share more soon. The technical report is coming shortly — stay tuned and stay in touch. 👀 #ToposLite #3DGS #3DGeneration #SpatialAI #AIGC #photograph #3DArtwork

InSpatio｜影溯

23,176 次观看 • 2 个月前

We release a technical blog on training PocketTTS, our 100M parameter on-device TTS, through drifting, a recent one-step generative objective from Deng et al. Less than 1% WER, high quality and voice cloning. To the best of our knowledge, it's the first speech model, and the first autoregressive model, trained this way.

We release a technical blog on training PocketTTS, our 100M parameter on-device TTS, through drifting, a recent one-step generative objective from Deng et al. Less than 1% WER, high quality and voice cloning. To the best of our knowledge, it's the first speech model, and the first autoregressive model, trained this way.

kyutai

17,569 次观看 • 3 天前

We got a robot to clean up homes that were never seen in its training data! Our new model, π-0.5, aims to tackle open-world generalization. We took our robot into homes that were not in the training data and asked it to clean kitchens and bedrooms. More below⤵️

We got a robot to clean up homes that were never seen in its training data! Our new model, π-0.5, aims to tackle open-world generalization. We took our robot into homes that were not in the training data and asked it to clean kitchens and bedrooms. More below⤵️

Physical Intelligence

537,029 次观看 • 1 年前

Introducing 👀Stereo4D👀 A method for mining 4D from internet stereo videos. It enables large-scale, high-quality, dynamic, *metric* 3D reconstructions, with camera poses and long-term 3D motion trajectories. We used Stereo4D to make a dataset of over 100k real-world 4D scenes.

Introducing 👀Stereo4D👀 A method for mining 4D from internet stereo videos. It enables large-scale, high-quality, dynamic, metric 3D reconstructions, with camera poses and long-term 3D motion trajectories. We used Stereo4D to make a dataset of over 100k real-world 4D scenes.

Linyi Jin

94,683 次观看 • 1 年前

🤔What does it take to train an omni world model from scratch? 📽️Petabytes of video. 6⃣pipeline stages. And every improvement in data quality pays off twice when you're training a model that both generates and understands video. Our Reka Labs' data team on how it works → #PhysicalAI #WorldModels #AI #RekaLabs

🤔What does it take to train an omni world model from scratch? 📽️Petabytes of video. 6⃣pipeline stages. And every improvement in data quality pays off twice when you're training a model that both generates and understands video. Our Reka Labs' data team on how it works → #PhysicalAI #WorldModels #AI #RekaLabs

Reka

15,279 次观看 • 2 个月前

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 次观看 • 2 年前

This is an update on Loftia’s Beta and Early Access timelines 🌱☀️ It’s been an incredible 3 years and we truly wouldn’t be here without the support from this amazing community! Thank you for everything, and we can’t wait to share our cozy world with you soon 💛

This is an update on Loftia’s Beta and Early Access timelines 🌱☀️ It’s been an incredible 3 years and we truly wouldn’t be here without the support from this amazing community! Thank you for everything, and we can’t wait to share our cozy world with you soon 💛

☁️ Loftia - Wishlist on Steam! ☁️

13,840 次观看 • 9 个月前

🚨 JUST IN: You’ve asked for it, and it’s happening! RavenQuest will be coming to MOBILE for Full-Release, making it more accessible for players around the world! 🌍 It’s still early, but we couldn’t wait to share this with you. Check out what we’ve got so far 👇

🚨 JUST IN: You’ve asked for it, and it’s happening! RavenQuest will be coming to MOBILE for Full-Release, making it more accessible for players around the world! 🌍 It’s still early, but we couldn’t wait to share this with you. Check out what we’ve got so far 👇

RavenQuest

21,504 次观看 • 2 年前

📢📢 𝐏𝐞𝐫𝐜𝐇𝐞𝐚𝐝: 𝐏𝐞𝐫𝐜𝐞𝐩𝐭𝐮𝐚𝐥 𝐇𝐞𝐚𝐝 𝐌𝐨𝐝𝐞𝐥 𝐟𝐨𝐫 𝐒𝐢𝐧𝐠𝐥𝐞-𝐈𝐦𝐚𝐠𝐞 𝟑𝐃 𝐇𝐞𝐚𝐝 𝐑𝐞𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧 & 𝐄𝐝𝐢𝐭𝐢𝐧𝐠📢📢 PercHead reconstructs realistic 3D heads from a single image and enables disentangled 3D editing via geometric controls and style inputs from images or text. At its core is a generalized 3D head decoder trained with perceptual supervision from DINOv2 and SAM 2.1. We find that our new perceptual loss formulation improves reconstruction fidelity compared to commonly-used methods such as LPIPS. Our trained reconstruction model is able to generate 3D-consistent heads from a single input image. Even with challenging side-view inputs, the model robustly infers missing regions for a coherent, high-fidelity output. In addition, our architecture seamlessly adapts to downstream tasks: by swapping the encoder, we can transform the model into a disentangled 3D editing pipeline. In this scenario, we can control geometry through - potentially hand-drawn - segmentation maps, and condition style via image or text prompt. We also provide an interactive GUI to enable the exploration of our editing pipeline. 🌍 📽️ Great work by Antonio Oroz and Tobias Kirschstein

📢📢 𝐏𝐞𝐫𝐜𝐇𝐞𝐚𝐝: 𝐏𝐞𝐫𝐜𝐞𝐩𝐭𝐮𝐚𝐥 𝐇𝐞𝐚𝐝 𝐌𝐨𝐝𝐞𝐥 𝐟𝐨𝐫 𝐒𝐢𝐧𝐠𝐥𝐞-𝐈𝐦𝐚𝐠𝐞 𝟑𝐃 𝐇𝐞𝐚𝐝 𝐑𝐞𝐜𝐨𝐧𝐬𝐭𝐫𝐮𝐜𝐭𝐢𝐨𝐧 & 𝐄𝐝𝐢𝐭𝐢𝐧𝐠📢📢 PercHead reconstructs realistic 3D heads from a single image and enables disentangled 3D editing via geometric controls and style inputs from images or text. At its core is a generalized 3D head decoder trained with perceptual supervision from DINOv2 and SAM 2.1. We find that our new perceptual loss formulation improves reconstruction fidelity compared to commonly-used methods such as LPIPS. Our trained reconstruction model is able to generate 3D-consistent heads from a single input image. Even with challenging side-view inputs, the model robustly infers missing regions for a coherent, high-fidelity output. In addition, our architecture seamlessly adapts to downstream tasks: by swapping the encoder, we can transform the model into a disentangled 3D editing pipeline. In this scenario, we can control geometry through - potentially hand-drawn - segmentation maps, and condition style via image or text prompt. We also provide an interactive GUI to enable the exploration of our editing pipeline. 🌍 📽️ Great work by Antonio Oroz and Tobias Kirschstein

Matthias Niessner

18,903 次观看 • 11 个月前

Today, we are open-sourcing Hunyuan World 1.1 (WorldMirror), a universal feed-forward 3D reconstruction model. 🚀🚀🚀 While our previously released Hunyuan World 1.0 (open-sourced, lite version deployable on consumer GPUs) focused on generating 3D worlds from text or single-view images, Hunyuan World 1.1 significantly expands the input scope by unlocking video-to-3D and multi-view-to-3D world creation. Highlights: 🔹Any Input, Maximized Flexibility and Fidelity: Flexibly integrates diverse geometric priors (camera poses, intrinsics, depth maps) to resolve structural ambiguities and ensure geometrically consistent 3D outputs. 🔹Any Output, SOTA Results：This elegant architecture simultaneously generates multiple 3D representations: dense point clouds, multi-view depth maps, camera parameters, surface normals, and 3D Gaussian Splattings. 🔹Single-GPU & Fast Inference: As an all-in-one, feed-forward model, Hunyuan World 1.1 runs on a single GPU and delivers all 3D attributes in a single forward pass, within seconds. 🌐Project Page: 🔗Github： 🤗Hugging Face： ✨Demo: 📄Technical Report:

Today, we are open-sourcing Hunyuan World 1.1 (WorldMirror), a universal feed-forward 3D reconstruction model. 🚀🚀🚀 While our previously released Hunyuan World 1.0 (open-sourced, lite version deployable on consumer GPUs) focused on generating 3D worlds from text or single-view images, Hunyuan World 1.1 significantly expands the input scope by unlocking video-to-3D and multi-view-to-3D world creation. Highlights: 🔹Any Input, Maximized Flexibility and Fidelity: Flexibly integrates diverse geometric priors (camera poses, intrinsics, depth maps) to resolve structural ambiguities and ensure geometrically consistent 3D outputs. 🔹Any Output, SOTA Results：This elegant architecture simultaneously generates multiple 3D representations: dense point clouds, multi-view depth maps, camera parameters, surface normals, and 3D Gaussian Splattings. 🔹Single-GPU & Fast Inference: As an all-in-one, feed-forward model, Hunyuan World 1.1 runs on a single GPU and delivers all 3D attributes in a single forward pass, within seconds. 🌐Project Page: 🔗Github： 🤗Hugging Face： ✨Demo: 📄Technical Report:

Tencent Hy

168,848 次观看 • 11 个月前

AI is only as good as its data. Everyone talks about building smarter AI models, but fewer people ask where the training data comes from. That’s what makes Humyn Labs interesting. They’re focused on collecting high-quality, real-world human data across different environments, behaviors, and sensory inputs. Because if AI is going to understand the physical world, it first needs to learn from the people who already live and work in it. I shared my thoughts in this short video. Let me know what stood out to you. 👇🎥 Humyn Labs KGeN Community #KGeN

AI is only as good as its data. Everyone talks about building smarter AI models, but fewer people ask where the training data comes from. That’s what makes Humyn Labs interesting. They’re focused on collecting high-quality, real-world human data across different environments, behaviors, and sensory inputs. Because if AI is going to understand the physical world, it first needs to learn from the people who already live and work in it. I shared my thoughts in this short video. Let me know what stood out to you. 👇🎥 Humyn Labs KGeN Community #KGeN

Bahdfella™ ❤️‍🔥

21,966 次观看 • 2 个月前

Can 1,000 real-world 3D scenes make a state-of-the-art model better? At OVER Research, we tested it by fine-tuning R3, a state-of-the-art streaming 3D reconstruction model, on roughly 1,000 scenes from the OVER dataset. The result: 25% lower trajectory error on our real-world environments compared to the deployed public checkpoint. And we did it using only 1,000 scenes, the same open dataset anyone can download for free, out of the 285,000+ real-world 3D scenes OVER has today. Read the full OVER Research publication: Download the 1,000-scene dataset on Hugging Face:

Can 1,000 real-world 3D scenes make a state-of-the-art model better? At OVER Research, we tested it by fine-tuning R3, a state-of-the-art streaming 3D reconstruction model, on roughly 1,000 scenes from the OVER dataset. The result: 25% lower trajectory error on our real-world environments compared to the deployed public checkpoint. And we did it using only 1,000 scenes, the same open dataset anyone can download for free, out of the 285,000+ real-world 3D scenes OVER has today. Read the full OVER Research publication: Download the 1,000-scene dataset on Hugging Face:

Over the Reality 🌐

34,753 次观看 • 1 个月前

📢📢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

18,351 次观看 • 4 个月前

World models hold a lot of promise for robotics, but they're data hungry and often struggle with long horizons. We learn models from a few (< 10) human demos that enable a robot to plan in completely novel scenes! Our key idea is to model *symbols* not pixels 👇

World models hold a lot of promise for robotics, but they're data hungry and often struggle with long horizons. We learn models from a few (< 10) human demos that enable a robot to plan in completely novel scenes! Our key idea is to model symbols not pixels 👇

Nishanth Kumar

83,024 次观看 • 1 年前

We have officially submitted our project to the Pump.fun Hackathon. Through this initiative, our goal is to test Grok in real world conditions and push it to its full operational limits. We want to explore the complete range of it's capabilities, observe how it performs under continuous and complex decision making scenarios, and clearly understand where it's strengths end and it's boundaries begin. This is not a theoretical experiment. We are applying Grok across hardware, software, and physical systems, using live data and real world constraints. Our long term vision is to scale this into something much bigger a fully automated, real world farm powered by robotics, intelligent systems, and AI driven optimization. By building in public and testing without shortcuts, we aim to learn fast, iterate openly, and discover how far this technology can truly go.

We have officially submitted our project to the Pump.fun Hackathon. Through this initiative, our goal is to test Grok in real world conditions and push it to its full operational limits. We want to explore the complete range of it's capabilities, observe how it performs under continuous and complex decision making scenarios, and clearly understand where it's strengths end and it's boundaries begin. This is not a theoretical experiment. We are applying Grok across hardware, software, and physical systems, using live data and real world constraints. Our long term vision is to scale this into something much bigger a fully automated, real world farm powered by robotics, intelligent systems, and AI driven optimization. By building in public and testing without shortcuts, we aim to learn fast, iterate openly, and discover how far this technology can truly go.

Grok Potato

32,252 次观看 • 8 个月前

🎉 LOFTIA IS COMING THIS FALL! 🎉 After 4 years of building Loftia, we’re so excited to announce our release window of Fall 2026! Loftia is a life-sim MMO where you can farm, craft, decorate and build a solarpunk world together! We can’t wait to see you in the skies soon 💛🥰🌱

🎉 LOFTIA IS COMING THIS FALL! 🎉 After 4 years of building Loftia, we’re so excited to announce our release window of Fall 2026! Loftia is a life-sim MMO where you can farm, craft, decorate and build a solarpunk world together! We can’t wait to see you in the skies soon 💛🥰🌱

☁️ Loftia - Wishlist on Steam! ☁️

181,381 次观看 • 4 个月前

"The Covid pandemic showed us the importance of training Africa’s health workforce. New health challenges have continued to arise, since then. We need to lean into advancements in science and technology, and provide better learning opportunities for health professionals. In Rwanda, we aim to quadruple our health workers in the next four years. The WHO Academy will be instrumental in helping us achieve this objective, and we intend to make full use of it. We owe it to our people to deliver consistent, affordable, high-quality health care. And for us, this means creating partnerships, where we can benefit from knowledge transfer, in essential fields like biotechnology." Today, President Kagame virtually addressed the inauguration of World Health Organization (WHO) Academy in Lyon, France.

"The Covid pandemic showed us the importance of training Africa’s health workforce. New health challenges have continued to arise, since then. We need to lean into advancements in science and technology, and provide better learning opportunities for health professionals. In Rwanda, we aim to quadruple our health workers in the next four years. The WHO Academy will be instrumental in helping us achieve this objective, and we intend to make full use of it. We owe it to our people to deliver consistent, affordable, high-quality health care. And for us, this means creating partnerships, where we can benefit from knowledge transfer, in essential fields like biotechnology." Today, President Kagame virtually addressed the inauguration of World Health Organization (WHO) Academy in Lyon, France.

Presidency | Rwanda

54,501 次观看 • 1 年前