Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

📢ScanEdit: Hierarchically-Guided Functional 3D Scan Editing Edit complex, real-world 3D scans with text -- Mohamed El Amine Boudjoghra combines LLM reasoning with geometric optimization to produce physically plausible, instruction-aligned scene edits Check it out:

angelaqdai's profile picture

Angela Dai

9,467 subscribers

12,984 görüntüleme • 1 yıl önce •via X (Twitter)

EğitimBilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

3D editing is hard: you need to ground an image + instruction and generate a faithful 3D shape in one forward pass -- no test-time optimization. So, we steer pretrained image-to-3D representations to do text-guided 3D edits; no massive 3D edit-pair dataset needed. Key trap: the “no-edit” solution is a nasty local minimum. We fix it with preference optimization, pushing the model to actually edit. Steer3D is the second work that adapts alignment ideas from LLMs to the 3D modality. SAM 3D also used DPO to improve its 3D generations.
0:22

3D editing is hard: you need to ground an image + instruction and generate a faithful 3D shape in one forward pass -- no test-time optimization. So, we steer pretrained image-to-3D representations to do text-guided 3D edits; no massive 3D edit-pair dataset needed. Key trap: the “no-edit” solution is a nasty local minimum. We fix it with preference optimization, pushing the model to actually edit. Steer3D is the second work that adapts alignment ideas from LLMs to the 3D modality. SAM 3D also used DPO to improve its 3D generations.

Georgia Gkioxari

115,844 görüntüleme • 5 ay önce

FirePlace Geometric Refinements of LLM Common Sense Reasoning for 3D Object Placement
0:45

FirePlace Geometric Refinements of LLM Common Sense Reasoning for 3D Object Placement

AK

49,980 görüntüleme • 1 yıl önce

📢ProcGen3D: Learning Neural Procedural Graphs for Image-to-3D Reconstruction Xinyi Zhang learns neural procedural graphs to generate high-fidelity 3D - MCTS-guided sampling maintains consistency with the input image, even from real images! Check it out:
0:38

📢ProcGen3D: Learning Neural Procedural Graphs for Image-to-3D Reconstruction Xinyi Zhang learns neural procedural graphs to generate high-fidelity 3D - MCTS-guided sampling maintains consistency with the input image, even from real images! Check it out:

Angela Dai

14,581 görüntüleme • 6 ay önce

Wild. Single shot text to 3D scene (gaussian splat) with no editing. This is approaching a usable scene asset for a game / 3D application.
0:21

Wild. Single shot text to 3D scene (gaussian splat) with no editing. This is approaching a usable scene asset for a game / 3D application.

martin_casado

56,136 görüntüleme • 10 ay önce

🕹️We are excited to introduce "ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation" ChronoEdit reframes image editing as a video generation task to encourage temporal consistency. It leverages a temporal reasoning stage that denoises with “video reasoning tokens” to "reason" on physically plausible edits. See the attached video for results. Project Page: Arxiv: Code and model are coming.
1:38

🕹️We are excited to introduce "ChronoEdit: Towards Temporal Reasoning for Image Editing and World Simulation" ChronoEdit reframes image editing as a video generation task to encourage temporal consistency. It leverages a temporal reasoning stage that denoises with “video reasoning tokens” to "reason" on physically plausible edits. See the attached video for results. Project Page: Arxiv: Code and model are coming.

Huan Ling

36,802 görüntüleme • 8 ay önce

🥁 The winners of the Google Maps Platform Awards are now live, and the Grand Prize goes to Innobrix! 🎉 The 3D configuration platform combines BIM data with Google Maps to provide real-world context for developments. 📢 Check it out along with the rest of our winners ⬇️
1:58

🥁 The winners of the Google Maps Platform Awards are now live, and the Grand Prize goes to Innobrix! 🎉 The 3D configuration platform combines BIM data with Google Maps to provide real-world context for developments. 📢 Check it out along with the rest of our winners ⬇️

Google Maps Platform

14,361 görüntüleme • 9 ay önce

Nvidia presents LATTE3D Large-scale Amortized Text-To-Enhanced 3D Synthesis Recent text-to-3D generation approaches produce impressive 3D results but require time-consuming optimization that can take up to an hour per prompt [21, 39]. Amortized methods like ATT3D [26]
0:31

Nvidia presents LATTE3D Large-scale Amortized Text-To-Enhanced 3D Synthesis Recent text-to-3D generation approaches produce impressive 3D results but require time-consuming optimization that can take up to an hour per prompt [21, 39]. Amortized methods like ATT3D [26]

AK

76,965 görüntüleme • 2 yıl önce

📢 Intrinsic Image Fusion for Multi-View 3D Material Reconstruction 📢 We combine generative material priors with inverse path tracing: 1) define a parametric texture space 2) fuse monocular predictions across views into consistent textures 3) optimize low-dimensional parameters for physically-grounded reconstructions. The results are relightable PBR textures for 3D scenes: check out the result on a real-world 3D scan from the ScanNet++ dataset! 🌍 🎥 Great work by Peter Kocsis Lukas Höllein!
0:31

📢 Intrinsic Image Fusion for Multi-View 3D Material Reconstruction 📢 We combine generative material priors with inverse path tracing: 1) define a parametric texture space 2) fuse monocular predictions across views into consistent textures 3) optimize low-dimensional parameters for physically-grounded reconstructions. The results are relightable PBR textures for 3D scenes: check out the result on a real-world 3D scan from the ScanNet++ dataset! 🌍 🎥 Great work by Peter Kocsis Lukas Höllein!

Matthias Niessner

19,833 görüntüleme • 5 ay önce

🔥 3D-LLMs go brrrr! 🚀 Excited to announce our latest research on scaling 3D-LLM training data to *million-scale* with *dense grounding*. 🌟 Introducing 3D-GRAND: a pioneering dataset featuring 40,087 household scenes paired with 6.2 million densely-grounded 3D-text pairs. 🏠💬 #LLM #3D #AI #ML #Robotics #EmbodiedAI
1:00

🔥 3D-LLMs go brrrr! 🚀 Excited to announce our latest research on scaling 3D-LLM training data to *million-scale* with *dense grounding*. 🌟 Introducing 3D-GRAND: a pioneering dataset featuring 40,087 household scenes paired with 6.2 million densely-grounded 3D-text pairs. 🏠💬 #LLM #3D #AI #ML #Robotics #EmbodiedAI

Jianing “Jed” Yang

27,626 görüntüleme • 2 yıl önce

LaRI Layered Ray Intersections for Single-view 3D Geometric Reasoning
0:45

LaRI Layered Ray Intersections for Single-view 3D Geometric Reasoning

AK

23,549 görüntüleme • 1 yıl önce

📢 Matrix-3D: Omnidirectional 3D World Generation Ever wanted to turn a single image or a text prompt into a massive, explorable 3D world? That’s exactly what Matrix-3D does.
0:42

📢 Matrix-3D: Omnidirectional 3D World Generation Ever wanted to turn a single image or a text prompt into a massive, explorable 3D world? That’s exactly what Matrix-3D does.

Skywork

30,331 görüntüleme • 9 ay önce

Using the new WordPress Command Palette to call an assistant that adds LLM generated text to a 3D world using natural language commands! "add text: write a short poem about the metaverse" This extends to image, audio and 3D objects in the future. WebXR holodeck style editing!
1:49

Using the new WordPress Command Palette to call an assistant that adds LLM generated text to a 3D world using natural language commands! "add text: write a short poem about the metaverse" This extends to image, audio and 3D objects in the future. WebXR holodeck style editing!

XR Publisher

17,891 görüntüleme • 2 yıl önce

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:
1:54

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

17,858 görüntüleme • 6 ay önce

Phones with 3D cameras to scan your face
4:30

Phones with 3D cameras to scan your face

Not A Number

42,613 görüntüleme • 10 ay önce

Semantically annotating 3D gaussian splats on the fly using gemini 3.1 + sparkjs 1. Load any 3D scene and hit scan 2. Get 2D detections from VLM 3. Cluster outputs & project into 3D world space 4. Save as a persistent 3D semantic layer Inspired by Alexander Chen's experiments with gemini visual intelligence. Just had to try to lift it from 2D to 3D!
0:21

Semantically annotating 3D gaussian splats on the fly using gemini 3.1 + sparkjs 1. Load any 3D scene and hit scan 2. Get 2D detections from VLM 3. Cluster outputs & project into 3D world space 4. Save as a persistent 3D semantic layer Inspired by Alexander Chen's experiments with gemini visual intelligence. Just had to try to lift it from 2D to 3D!

Bilawal Sidhu

56,125 görüntüleme • 23 gün önce

Introducing Stable Virtual Camera: This multi-view diffusion model transforms 2D images into immersive 3D videos with realistic depth and perspective—without complex reconstruction or scene-specific optimization.
0:39

Introducing Stable Virtual Camera: This multi-view diffusion model transforms 2D images into immersive 3D videos with realistic depth and perspective—without complex reconstruction or scene-specific optimization.

Stability AI

197,203 görüntüleme • 1 yıl önce

3D AI is leveling up! Rodin 3D AI can create stunning, high-quality 3D models from just text or image inputs. And with its latest update, it can even generate 8K HDRI textures to bring your models to life. Check out the link in the comments!
1:03

3D AI is leveling up! Rodin 3D AI can create stunning, high-quality 3D models from just text or image inputs. And with its latest update, it can even generate 8K HDRI textures to bring your models to life. Check out the link in the comments!

el.cine

46,010 görüntüleme • 1 yıl önce

Text-to-3D with Blocks📦 Striving towards real-time creation on #Rodin Gen-1 with multiple 3D ControlNets! 🚀 Rodin Gen-1 will be the largest ever 3D large model, and we are planing to open it to public🎉 Stay tuned!
1:22

Text-to-3D with Blocks📦 Striving towards real-time creation on #Rodin Gen-1 with multiple 3D ControlNets! 🚀 Rodin Gen-1 will be the largest ever 3D large model, and we are planing to open it to public🎉 Stay tuned!

Hyper3D by Deemos

41,161 görüntüleme • 2 yıl önce

2/ 3D world-building with NVIDIA NVIDIA used Edify AI to create a detailed 3D desert in minutes during a live demo Edify 3D generates editable 3D meshes from text or image prompts
0:52

2/ 3D world-building with NVIDIA NVIDIA used Edify AI to create a detailed 3D desert in minutes during a live demo Edify 3D generates editable 3D meshes from text or image prompts

Poonam Soni

107,014 görüntüleme • 1 yıl önce

3D Gaussian Splats as an art form? 🎨 Beautiful scan by Eric Paré. Check it out on SuperSplat:
0:33

3D Gaussian Splats as an art form? 🎨 Beautiful scan by Eric Paré. Check it out on SuperSplat:

Will Eastcott

14,905 görüntüleme • 1 yıl önce