跳过导航
Introducing Marigold 🌼 - a universal monocular depth estimator, delivering incredibly sharp…
0:00 / 0:36

Introducing Marigold 🌼 - a universal monocular depth estimator, delivering incredibly sharp…

Anton Obukhov

@AntonObukhov12702 位关注者

在 X 上查看
1,421 个赞下载原帖

490,940 次观看

Introducing Marigold 🌼 - a universal monocular depth estimator, delivering incredibly sharp predictions in the wild! Based on Stable Diffusion, it is trained with synthetic depth data only and excels in zero-shot adaptation to real-world imagery. Check it out: 🌐 Website: 🤗 Hugging Face Space: 📄 Paper: 👾 Code: The team: Bingxin Ke (KeBingxin), yours truly (Anton Obukhov), Shengyu Huang (Shengyu Huang), Nando Metzger (Nando Metzger), Rodrigo Caye Daudt (Rodrigo Daudt), and Konrad Schindler. #ComputerVision #PRS #ETHZurich

9 条评论

评论来自 X (Twitter) 上的原帖

  • AI葵(Aoi) 🇹🇼NeRF技術型Vtuber 的头像

    @kwea123

    70s of inference for an image is very slow..

    9 个赞

  • Anton Obukhov 的头像

    @AntonObukhov1

    We focus on the quality first - speed will come later through research and improvements in diffusion samplers. Arguably, Stable Diffusion is also slower than most of the GANs :)

    42 个赞

  • Bilawal Sidhu 的头像

    @bilawalsidhu

    Damn those are some crispy depth maps!

    7 个赞

  • Chris Offner 的头像

    @chrisoffner3d

    Wow, those depth maps look amazing. Great work! There goes my course project for Deep Learning. 😅

    5 个赞

  • Boyuan Chen 的头像

    @BoyuanChen0

    Is this metric or relative depth?

    4 个赞

  • Anton Obukhov 的头像

    @AntonObukhov1

    Relative

    6 个赞

  • Janek Mann 的头像

    @janekm

    I also consider this strong evidence that Stable Diffusion learned much stronger 3D scene understanding than commonly assumed.

    4 个赞

  • Omar Sanseviero 的头像

    @osanseviero

    Very very cool!

    3 个赞

  • Tom 的头像

    @Tom14985282

    Wow. More detailed than the ground truth, very cool to see.

    3 个赞