VAST AI Research (@VastAIResearch) / X

VAST AI Research

111 posts

VAST AI Research

@VastAIResearch

VAST AI Research advances the state-of-the-art in 3D + Generative AI.

San Francisco, CA

Joined November 2023

Pinned
VAST AI Research
@VastAIResearch
Jun 1
Open-sourcing TripoSplat - the first 3D Gaussian generator that learns to smartly allocate Gaussians from data and supports outputting an arbitrary number of Gaussians. The technique behind it, Density-Sampled Gaussians (DeG), has been accepted to @siggraph 2026.
00:00
Made with AI
44K
VAST AI Research
@VastAIResearch
Jul 7, 2024
#SIGGRAPH2024 is just around the corner! We’re excited to present 𝘾𝙝𝙖𝙧𝙖𝙘𝙩𝙚𝙧𝙂𝙚𝙣, our latest work that transforms a single image into 3D pose-unified character meshes with high-quality, consistent appearances. Great for downstream rigging and animation workflows. 👇
00:00
90K
VAST AI Research
@VastAIResearch
Dec 1, 2024
🎉 Our team is thrilled to be at #SIGGRAPHAsia2024 in Tokyo this week! Join us for: 📑 TEXGen - Technical Papers (Wed PM) 🎨 Tripo Doodle - Real-Time Live! (Fri PM) 🤝 Visit us at Booth S-05 (Wed-Fri) for exciting demos and surprises! Details ⬇️
00:00
15K
VAST AI Research
@VastAIResearch
Mar 29, 2025
We're excited to release TripoSG & TripoSF—breakthrough open 3D GenAI models from Tripo @tripoai. TripoSG excels at image-to-mesh generation; TripoSF redefines high-res, complex topology with SparseFlex. Now OPEN-SOURCE!
00:00
5K
VAST AI Research
@VastAIResearch
Jul 7, 2024
Replying to @VastAIResearch
Check out the project page and explore the code: - Project page: charactergen.github.io - Code: github.com/zjp-shadow/Cha… Huge thanks to the team for their incredible effort!
5.5K
VAST AI Research
@VastAIResearch
Jan 16, 2024
Thanks @_akhaliq for sharing CSD (Classifier Score Distillation), and our work got accepted at @iclr_conf! Check out the project here: xinyu-andy.github.io/Classifier-Sco… #ICLR2024
AK
@_akhaliq
Oct 31, 2023
Text-to-3D with classifier score distillation paper page: huggingface.co/papers/2310.19… Text-to-3D generation has made remarkable progress recently, particularly with methods based on Score Distillation Sampling (SDS) that leverages pre-trained 2D diffusion models. While the usage of
15K
VAST AI Research
@VastAIResearch
Jul 11, 2024
The gradio demo is online:
VAST AI Research
@VastAIResearch
Jul 7, 2024
#SIGGRAPH2024 is just around the corner! We’re excited to present 𝘾𝙝𝙖𝙧𝙖𝙘𝙩𝙚𝙧𝙂𝙚𝙣, our latest work that transforms a single image into 3D pose-unified character meshes with high-quality, consistent appearances. Great for downstream rigging and animation workflows. 👇
00:00
CharacterGen - a Hugging Face Space by VAST-AI
From huggingface.co
27K
VAST AI Research
@VastAIResearch
Feb 27, 2024
Thrilled to share that 𝗮𝗹𝗹 (𝟲/𝟲) submissions from our VAST AI Research team made it to #CVPR2024! Covering 𝗧𝗚𝗦, 𝗣𝗜𝟯𝗗, 𝗦𝗖-𝗚𝗦, 𝗪𝗼𝗻𝗱𝗲𝗿𝟯𝗗, 𝗘𝗽𝗶𝗗𝗶𝗳𝗳, and 𝗗𝗿𝗲𝗮𝗺𝗖𝗼𝗺𝗽𝗼𝘀𝗲𝗿. Kudos to the team & collaborators! Details: (1/n)
5.8K
VAST AI Research
@VastAIResearch
Apr 2, 2024
Excited to introduce LD-T3D, our leap in text-based 3D model retrieval. With 100K text-to-3D model pairs and 89K distinct models from Objaverse, this benchmark sets a new standard for diversity and scale.
Gradio
@Gradio
Apr 1, 2024
🌐𝐋𝐃-𝐓𝟑𝐃 🔍Discover the power of text-based 3D model retrieval with LD-T3D! 📊This large-scale & diverse benchmark introduces 100K text-to-3D model pairs, featuring 89K distinct 3D models from Objaverse and 1,000 descriptive text queries. 👇Learn More
00:00
13K
VAST AI Research
@VastAIResearch
Jan 27, 2025
🚀 Tripo v2.5 is LIVE, and we’re thrilled to share what this leap forward means for 3D creators.
00:00
00:43
Tripo
@tripoai
Jan 26, 2025
Tripo v2.5 is here! 🎉 This major update made significant enhancements to our algorithm, enabling our creators to achieve exceptional levels of detail and geometric precision in their models. #3dmodeling #tripoai #tripo #3d #3Dprinting
4.6K
VAST AI Research
@VastAIResearch
Dec 3, 2024
We're honored to share that TEXGen received the Best Paper Honorable Mention at #SIGGRAPHAsia2024! A large DM to enable direct texture synthesis in UV space - a new direction for high-quality 3D content creation. Come join @andy_yx27’s presentation: Wed 4:30-5:40pm
VAST AI Research
@VastAIResearch
Dec 1, 2024
Replying to @VastAIResearch
TEXGen: A large diffusion model learns directly in UV texture space, enabling text/image-guided texture synthesis, inpainting & completion at high resolution. 🔗 arxiv.org/abs/2411.14740 🌐 cvmi-lab.github.io/TEXGen/ Wed 4:30-5:40pm , "Text, Texturing, and Stylization" Session.
9.1K
VAST AI Research
@VastAIResearch
Jan 18, 2024
Thanks @_akhaliq for featuring our work! TGS (Triplane meets Gaussian Splatting) proposes a new hybrid 3D representation and high-quality 3D reconstruction from single-view images within a second! 🚀 Try it now: github.com/VAST-AI-Resear…
AK
@_akhaliq
Jan 18, 2024
VAST AI Research released model for Triplane Meets Gaussian Splatting on Hugging Face Fast and Generalizable Single-View 3D Reconstruction with Transformers model: huggingface.co/VAST-AI/Tripla… demo: huggingface.co/spaces/VAST-AI…
00:00
8.5K
VAST AI Research
@VastAIResearch
Jul 2, 2024
Thrilled to share our UniDream is accepted to #ECCV2024! 🌟 Explore our relightable 3D asset generation with PBR materials: yg256li.github.io/UniDream/ Big kudos to the team!
AK
@_akhaliq
Dec 15, 2023
UniDream: Unifying Diffusion Priors for Relightable Text-to-3D Generation paper page: huggingface.co/papers/2312.08… Recent advancements in text-to-3D generation technology have significantly advanced the conversion of textual descriptions into imaginative well-geometrical and finely
00:00
2.8K
VAST AI Research
@VastAIResearch
Jan 17, 2024
The Gradio demo of TGS (Triplane meets Gaussian Splatting) is now trending on Hugging Face! 🔥🔥🔥
8.5K