Log inSign up
Xiao-Xiao Long
119 posts
user avatar
Xiao-Xiao Long
@xxlong0
Associate Professor @ NJU. | Organizer of @anysyn3d Spatial Intelligence & Robotics.
xxlong.site
Joined November 2023
260
Following
1,253
Followers
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 30, 2024
    Create your 3D portrait with only a single image using our powerful model Era3D. @ylecun Era3D is a new high-resolution cross-domain diffusion model, which produces 512x512 color images and normal maps (Wonder3D and Syncdreamer only support 256x256 resolution). With such 2D
    Image
    00:00
    Image
    00:22
    user avatar
    Gradio
    @Gradio
    May 29, 2024
    🚀 Era3D: A new AI model that creates high-res 🗿3D images from multiple viewpoints using just one input image!🌄 - Generates high-quality images up to 512×512 pixels🎯 - Uses efficient row-wise attention to reduce computation⚡ - 12x more efficient than sota methods💪
    98K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Jun 18, 2024
    Extract accurate depth from your 3D gaussians! Fast, accurate, and easy to use. Code of RaDe-GS is released! project page:baowenz.github.io/radegs/ code:github.com/BaowenZ/RaDe-GS paper:github.com/BaowenZ/RaDe-G… No need to change your 3D GS representations, we propose a play-and-plug
    Image
    00:00
    31K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 26, 2024
    An image + a detailed normal map generated by GeoWizard --> a relightable image. Geowizard is a generative geometry estimation model that could jointly produce high-quality depth and normal with intricate details. Try our huggingface demo here: huggingface.co/spaces/lemonad…
    Image
    00:00
    30K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Jun 19, 2024
    Metric3D V2. Currently it's hard for researcher to gain attention without significant promotions. Anyway recently we notice significant advancements in monodepth. However if you would like to obtain depth with true scale for slam or reconstruction tasks. TRY METRIC3D V2. Super
    Image
    00:00
    user avatar
    Dmytro Mishkin 🇺🇦
    @ducha_aiki
    Jun 18, 2024
    It looks like that monodepth community keeps ignoring Metric3D results, both v1 and v2. Any ideas why? #CVPR2024
    24K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Mar 25, 2024
    The huggingface gradio demo of GeoWizard is out! Single click for both depth and normal with rich details. huggingface.co/spaces/lemonad… The reconstruction feature will be soon. Visit our page fuxiao0719.github.io/projects/geowi… for more comparisons. #Depth #Normal #Gradio #Hugggingface
    Image
    58K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 6, 2024
    Happy to share our project GaussainPro, which is just accepted to ICML 2024. We propose a progressive propagation strategy to guide the densification of the 3D Gaussians. Compared to the simple split and clone strategies used in 3DGS, our method produces more reliable and
    Image
    00:00
    16K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Jul 8, 2024
    Our ECCV paper GeoWizard could produce high-quality depth and normal, yielding good 3D shapes. Check out our page: fuxiao0719.github.io/projects/geowi…
    Image
    00:00
    21K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 24, 2024
    Very excited to share our recent work CraftsMan3D. CraftsMan3D (aka 匠心) is a two-stage 3D native mesh diffusion model with a normal-based geometry refiner by mimicking the modeling workflow of artist/craftsman. @weiyuli99072112 The first stage is a large 3D native diffusion
    Image
    00:00
    31K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 29, 2024
    Thanks for the tweet. We improve 3D Gaussain splatting for dash cam videos that contain serious reflections and obstructions. With Dash cam videos, it's possible to leverage massive driving data. Visit our page at linhanwang.github.io/dcgaussian/ Codes will be soon.
    Image
    00:00
    Image
    Image
    Image
    user avatar
    Zhenjun Zhao
    @zhenjun_zhao
    May 29, 2024
    DC-Gaussian: Improving 3D Gaussian Splatting for Reflective Dash Cam Videos Linhan Wang, Kai Cheng, Shuo Lei, Shengkun Wang, Wei Yin, Chenyang Lei, @xxlong0, Chang-Tien Lu tl;dr: handle reflections & occlusions & illumination changes & rendering details arxiv.org/pdf/2405.17705
    44K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 24, 2024
    Have a try: huggingface.co/spaces/wyysf/C…
    Image
    00:00
    Image
    00:34
    user avatar
    Xiao-Xiao Long
    @xxlong0
    May 24, 2024
    Very excited to share our recent work CraftsMan3D. CraftsMan3D (aka 匠心) is a two-stage 3D native mesh diffusion model with a normal-based geometry refiner by mimicking the modeling workflow of artist/craftsman. @weiyuli99072112 The first stage is a large 3D native diffusion
    19K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Mar 10, 2024
    I just had a look at the cvpr24 accepted papers. There are nearly 70 papers named with “Gaussain”. They are about avatars, efficiency, surface reconstruction, dynamic, reflective surfaces, 3D generation, SLAM and so on. Happy but feel stressed about the crowded topic.
    51K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    May 29, 2024
    Checkout our new MV cross-domain diffusion model. Higher resolution than Wonder3D. (512 vs 256) More details at project page : penghtyx.github.io/Era3D/
    user avatar
    Gradio
    @Gradio
    May 29, 2024
    🚀 Era3D: A new AI model that creates high-res 🗿3D images from multiple viewpoints using just one input image!🌄 - Generates high-quality images up to 512×512 pixels🎯 - Uses efficient row-wise attention to reduce computation⚡ - 12x more efficient than sota methods💪
    Image
    00:00
    23K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Jul 1, 2024
    Very excited to share that GeoWizard is just accepted to ECCV 2024. Very appreaciate the efforts from every author, especially for Xiao Fu !
    user avatar
    Xiao-Xiao Long
    @xxlong0
    May 26, 2024
    An image + a detailed normal map generated by GeoWizard --> a relightable image. Geowizard is a generative geometry estimation model that could jointly produce high-quality depth and normal with intricate details. Try our huggingface demo here: huggingface.co/spaces/lemonad…
    Image
    00:00
    10K
  • user avatar
    Xiao-Xiao Long
    @xxlong0
    Mar 26, 2024
    I really realy like the Marigold work very very much. Because many people would like me to discuss about Geowizard and Marigold, so here are some information. When I was conducting project wonder3D, which uses diffusion models to produce mulit-view normals for 3D generation, I
    Image
    12K

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
Advertisement
Advertisement