Matthias Niessner

@MattNiessner

Professor for Visual Computing & Artificial Intelligence Co-Founder

Munich, Bavaria
niessnerlab.org/publications.h…
Joined March 2015
177 Photos and videos Photos and videos

Tweets

You blocked @MattNiessner

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @MattNiessner

  1. 3 hours ago

    Happening now :)

    Tomorrow in our TUM AI - Lecture, we have Chelsea Finn talking about "Broad Data for Broad Robot Generalization" Live Stream here: Starts at 6pm CEST / 9am PST! Come and join us!
    6 likes
    Undo
  2. Apr 29

    Tomorrow in our TUM AI - Lecture, we have Chelsea Finn talking about "Broad Data for Broad Robot Generalization" Live Stream here: Starts at 6pm CEST / 9am PST! Come and join us!

    2 retweets 40 likes
    Undo
  3. Apr 27

    Helping fact checkers to fight misinformation: our COSMOS work automatically detects out-of-context image use! just released the dataset of 200K images & 450K textual captions! Video Data

    10 retweets 22 likes
    Undo
  4. Apr 26

    We released the code for "SceneFormer: Indoor Scene Generation with Transformers" by and From a room layout or text input, our transformer model generates realistic indoor scenes! Code: Project:

    2 replies 50 retweets 224 likes
    Undo
  5. Retweeted
    Apr 23

    Excited to share our work Pri3D, learning 3D priors for 2D scene understanding tasks. Pre-training on 3D learns complementary features to ImageNet pre-trained model, leading to an 11.9% improvement on 20% data. w/ ben

    1 reply 21 retweets 99 likes
    Undo
  6. Apr 23

    Happening right now :)

    Tomorrow in our TUM AI - Lecture, we have talking about "Towards Virtual Humans: Putting Realistic People in Realistic Scenes doing Realistic Things" Live Stream here: Starts at 6pm CEST / 9am PST! Come and join us!
    3 likes
    Undo
  7. Apr 23

    How to successfully publish a paper at a top-tier venue like , , ? Key is to understand the psychology of your research area: what sells? what baselines are expected? This making it particularly hard for new researchers even with a great idea/method.

    6 replies 23 retweets 222 likes
    Undo
  8. Apr 22

    I might be giving up asking C++ questions... The ease of use of machine Learning frameworks seems to have killed actual coding :)

    3 replies 4 retweets 76 likes
    Undo
  9. Apr 22

    Tomorrow in our TUM AI - Lecture, we have talking about "Towards Virtual Humans: Putting Realistic People in Realistic Scenes doing Realistic Things" Live Stream here: Starts at 6pm CEST / 9am PST! Come and join us!

    4 replies 20 retweets 101 likes
    Undo
  10. Apr 21

    In 2020 we have seen several breakthroughs in neural rendering; still rapid progress happening right now! Question is what will 2021 be mostly about? Maybe self-supervised representation learning :) ? Already quite some exciting works out there :)

    4 replies 1 retweet 50 likes
    Undo
  11. Retweeted
    Apr 20

    What if creating a video was as easy as writing an email? Thrilled to partner with in building the future of synthetic video, and work with My thoughts on the investment 👇👇👇

    1 reply 4 retweets 37 likes
    Undo
  12. Retweeted
    Apr 20

    Incredibly proud to announce our $12.5m Series A, led by and 🥳 Humbled to be in the company of iconic co's like 🙏🏻 Last 9 months have been *crazy* Here's a recap & what's next... 👇

    8 replies 15 retweets 99 likes
    Show this thread
    Show this thread
    Undo
  13. Retweeted
    Apr 20

    1/ We're super excited to announce our $12.5M Series A - the largest investment in AI video space to date. The funding is led by , the fund behind some of the companies you might have heard of: , , 😉

    2 replies 13 retweets 38 likes
    Show this thread
    Show this thread
    Undo
  14. Apr 20

    Our startup Synthesia has raised a $12.5M Series A lead by ! So proud of the team - really amazing people to work with - impressive what they have accomplished in only a few years!

    3 replies 1 retweet 69 likes
    Undo
  15. Apr 18

    Roadmap back to normal lives: we need to reach 200 vaccine doses / 100 people: -> Israel 119/200 -> UK 60/200 -> US 58/200 -> Ger 24/200 Germany several months behind...we need to do better, seriously... Also worrying: Israel / UK progress plateauing.

    1 reply 1 retweet 19 likes
    Undo
  16. Apr 16

    Happening right now :)

    Tomorrow in our TUM AI - Lecture, we have talking about "Sensible Algorithms for Learning from Geometric Data" Live Stream here: Starts at 6pm CEST / 10am PST! (careful about the summer time shift in the US) Come and join us!
    6 likes
    Undo
  17. Apr 15

    Tomorrow in our TUM AI - Lecture, we have talking about "Sensible Algorithms for Learning from Geometric Data" Live Stream here: Starts at 6pm CEST / 10am PST! (careful about the summer time shift in the US) Come and join us!

    1 reply 10 retweets 63 likes
    Undo
  18. Apr 14

    . made a really nice project page about his Neural Parametric Models work: The latent space interpolation nicely illustrates the shape vs pose disentanglement :) Great work by

    3 retweets 17 likes
    Undo
  19. Apr 13

    The RTX 3090 *Turbo* was the only consumer RTX 3090 variant that could physically fit into deep learning machines -- it seems to have disappeared from the market. But guess no big deal, there are still the Quadro/Pro counterparts with similar hardware specs, right :) ?

    3 replies 1 retweet 9 likes
    Undo
  20. Retweeted
    Apr 13

    Check out our neural RGB-D surface reconstruction method which effectively incorporates depth measurements into a neural radiance field formulation resulting in more complete and cleaner meshes.

    4 replies 39 retweets 217 likes
    Undo

@MattNiessner hasn't Tweeted yet.

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·