Matthias Niessner

@MattNiessner

Professor for Visual Computing and Artificial Intelligence @ TU Munich

Vrijeme pridruživanja: ožujak 2015.

Tweetovi

Blokirali ste korisnika/cu @MattNiessner

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @MattNiessner

  1. 2. velj

    Live weather report by a photo-realistic AI-avatar! Weather data is pulled live from BBC API and videos are generated on-the-fly. Great collaboration by and :)

    Poništi
  2. proslijedio/la je Tweet
    Poništi
  3. 1. velj

    View outside my office - pretty mesmerizing, in particular, the lighting around the TUM tower :)

    Poništi
  4. 30. sij

    There's so many cool applications for 3D scanning, but we still need to work on making them photo-realistic while allowing for free view-point changes!

    Poništi
  5. 28. sij

    Always a good resource if you are looking for the top-tier / venues: Really interesting to see how the growth of the respective sub-areas - vision very dominant :)

    Poništi
  6. 27. sij

    Portland is amazing (the one on the east coast with lobster!!!)

    Poništi
  7. 27. sij

    We've released our ScanRefer Code & Dataset: 46,173 language descriptions of 9,943 objects in 3D scans! Learning Object detection in 3D scans from Language descriptions. Video: Paper:

    Poništi
  8. 25. sij

    Highlighting online course certificates (AndrewNg/Coursera et al.) in CVs for job apps in / has been a 100% correlation with little to no actual background knowledge. Currently, using it as pre-filter... Really wondering what went wrong there...Thoughts?

    Poništi
  9. proslijedio/la je Tweet
    24. sij

    Our Technical Papers jurors Alton Glass, , and reflect on their fondest SIGGRAPH memories and what excites them most about their work. Technical Paper submissions close TODAY!

    Poništi
  10. 23. sij

    Sad that computer science education at elementary-/high-school level is almost non-existent in pretty much all developed countries in the west. Every basic education *MUST* include programming skills! Economically so relevant; but almost no action :/

    Poništi
  11. 22. sij

    Check out Image-guided Neural Object Rendering by Combine benefits of image-based rendering and GAN-based image synthesis to generate photo-realistic re-renderings, including view-dependent effects :)

    Poništi
  12. 16. sij

    Check out the personalized AI-video generation demo by ! Looks pretty realistic :) -> Generate your own video avatar, personalized fitness buddy, customized ad, eLearning content, and may more :) -> Click generate Video!

    Poništi
  13. 14. sij

    All positions are fully funded; little to no teaching/bureaucratic overhead with the sole focus on cutting-edge publications at top-tier vision, ML, and graphics venues! We also have a fantastic, hard-working team within a great work environment :)

    Prikaži ovu nit
    Poništi
  14. 14. sij

    We have several openings for PhD Students, Visiting PhD Students, and PostDocs: Topics incl.: AI-based video/image generation, neural rendering, media forensics/forgery detection, 3D reconstruction, RGB-D scanning, scene understanding, deep learning in 3D

    Prikaži ovu nit
    Poništi
  15. 12. sij

    Surprising that many German/EU Universities still have W1/non-tenure-track professorship openings. Makes no sense – no career options but all the prof bureaucracy/overhead. Rather do PostDoc with academic freedom/high impact research. Time to make tenure-track default option!

    Poništi
  16. proslijedio/la je Tweet
    8. sij

    Unser Vorsitzender Alexander begrüßt , Professor an der TU München, der mit uns über Visual Computing und : „Autonomes Fahren, virtual Reality, ethische Fragen - das alles wird in Zukunft enorm an Bedeutung gewinnen“ 🖥️

    Poništi
  17. 7. sij

    Heading to - very happy to be an invited speaker talking about , , and - highlighting all the amazing opportunities! Education and research is the future of our country :) Thanks for the invite

    Poništi
  18. 24. pro 2019.

    I made myself a Christmas present and deleted all emails in my inbox - I'm free !!! :) Happy Holidays! P.S. if it was important, please re-ping next year :)

    Poništi
  19. 20. pro 2019.

    Happy to share our ScanRefer work: - ScanRefer learns to detect objects in 3D scenes from language descriptions - New dataset of 46,173 language descriptions of 9,943 objects in 3D scans Hope to see more work between NLP and 3D vision :)

    Poništi
  20. 13. pro 2019.

    Check out Neural Voice Puppetry :) Voice2Video and Text2Video! - can create photo-realistic videos from voice input - can also create videos from any text input Paper:

    Poništi

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·