Kevin Zakka

@kevin_zakka

CS MS . Previously robotics research .

Palo Alto, CA
Vrijeme pridruživanja: siječanj 2016.

Tweetovi

Blokirali ste korisnika/cu @kevin_zakka

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

  1. Prikvačeni tweet
    22. sij

    Our paper has been accepted to ICRA 2020! Looking forward to meeting and chatting with new and old faces there 🙂

    Poništi
  2. prije 14 sati

    Does anyone know how to get around YouTube scraping slowdowns and blocks? youtube-dl has stopped working after a few thousand videos.

    Poništi
  3. 2. velj

    Super cool, I've always wanted to test this hypothesis 😯 Checking for pairwise device distance with GPS probably won't fix it since it's smaller than GPS error rate? What about bluetooth pings?

    Poništi
  4. proslijedio/la je Tweet
    2. velj
    Odgovor korisnicima

    I might be biased but feature detection is one of my stronger suits. I'm just not fit to be tied.

    Poništi
  5. 1. velj

    Would be cool to pool bookmarks of favorite sites from a wide range of people (friends, experts, etc.). That way, you can get a curated version of the web from any topic you want to read about.

    Poništi
  6. 30. sij
    Poništi
  7. 30. sij

    I'll be joining X as an AI Resident this summer, working on the Everyday Robot Project. Super excited to make progress on teaching robots to be useful in our messy everyday lives.

    Poništi
  8. proslijedio/la je Tweet
    30. sij

    Love this! Google project sunroof. Just type in your address to know the solar potential of your roof.

    Poništi
  9. proslijedio/la je Tweet
    23. sij

    Interactive demo to see how IEEE floats are represented

    Poništi
  10. 19. sij

    It's that period of time in the research cycle 😅

    Poništi
  11. 18. sij

    "One takeaway is that results on small-scale experiments don’t always carry over to larger experiments." This was my exact experience for the importance sampling SGD experiments 😰

    Poništi
  12. proslijedio/la je Tweet
    18. sij

    New blog post on three of my recent ML research projects that _didn't_ pan out. Negative results can be informative, and I think the ideas are still interesting!

    Prikaži ovu nit
    Poništi
  13. 16. sij

    Would be nice if we could pin emails at the top of the inbox, make it happen!

    Poništi
  14. 15. sij

    Already halfway through January, crazy how fast time flies...

    Poništi
  15. proslijedio/la je Tweet
    14. sij

    2. Put yourself in the shoes of a master student who has to start from scratch with your code: - give them a ride up to the end with pre-trained models - focus examples/code on open-access datasets (not everybody can pay for CoNLL-2003)

    Prikaži ovu nit
    Poništi
  16. 15. sij

    Met Prof. Justin Solomon at Stanford today :) He was kind enough to sign my copy of his numerical linear algebra book 🤓

    Poništi
  17. proslijedio/la je Tweet
    13. sij

    I need a button in Google Hangouts that says "be right there, just untangling headphones".

    Poništi
  18. proslijedio/la je Tweet
    10. sij

    Check out our work on self-supervising keypoint and depth learning: - we lift keypoints to 3D and detect salient image regions with stable geometry. By training our system end-to-end we achieve SOTA matching and ego-motion estimation results (via DSO).

    Poništi
  19. proslijedio/la je Tweet
    10. sij

    Awesome ML collaboration led by Jie Li and her intern Rui Hou to make panoptic segmentation (instance+semantic) practical: single-stage, real-time, accurate, weakly-supervised! Main idea: a global self-attention module to reduce/reuse/recycle detections pre-NMS.

    Poništi
  20. proslijedio/la je Tweet
    9. sij

    What I did over my winter break! It gives me great pleasure to share this summary of some of our work in 2019, on behalf of all my colleagues at & .

    Prikaži ovu nit
    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:

    ·