Nikolaos Sarafianos

@sarafianosn

Research Scientist at Facebook Reality Labs working on 3D humans. Previously CS PhD from .

San Francisco, CA
nsarafianos.github.io
Vrijeme pridruživanja: prosinac 2016.
1 fotografija ili videozapis Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @sarafianosn

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

  1. Prikvačeni tweet
    22. srp 2019.

    I'm beyond excited to share that our paper with and Prof. Kakadiaris on "Adversarial Representation Learning for Text-Image Cross-Domain Matching" has been accepted to . Preprint coming soon.

    1 proslijeđeni tweet 11 korisnika označava da im se sviđa
    Poništi
  2. proslijedio/la je Tweet
    27. sij

    A fantastic new paper by Thomas Steinke and Lydia Zakynthinou ( and ). They use Conditional Mutual Information as a perspective to understand generalization, capturing VC dimension, compression schemes, differential privacy, & more.

    8 proslijeđenih tweetova 60 korisnika označava da im se sviđa
    Poništi
  3. proslijedio/la je Tweet
    14. sij

    BigGAN samples are famously photo-realistic but limited in diversity for some classes. Slightly modifying only the class embeddings (network unchanged) can reduce the diversity gap by ~50%! Work with Long Mai and led by fantastic !! Paper & video:

    67 proslijeđenih tweetova 243 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  4. proslijedio/la je Tweet
    10. pro 2019.

    Happy to share our recent work, "Geometric Capsule Autoencoders for 3D Point Clouds." The main idea is that instead of finding agreement among parts of an object, we find agreement among different views of the object.

    i
    68 proslijeđenih tweetova 268 korisnika označava da im se sviđa
    Poništi
  5. 9. pro 2019.

    Two very interesting works on 3D humans showed up on arxiv in the past 2 days: 1) Generating 3D People in Scenes without People () 2) CLOTH3D: Clothed 3D Humans ()

    Poništi
  6. proslijedio/la je Tweet
    6. pro 2019.

    Can backdoor attacks be successful without using incorrect labels? Yes, you just need to make poisoned inputs harder! Check out our work with and

    1 reply 18 proslijeđenih tweetova 75 korisnika označava da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    24. stu 2019.

    If you're interested in optical flow, check out Creative Flow+, a large-scale multi-style artistic video dataset (densely labeled with GT)! Awesome job by Masha Shugrina! Ziheng Liang Jiaman Li, Angad Singh, Karan Singh Webpage:

    1 reply 51 proslijeđeni tweet 244 korisnika označavaju da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    13. stu 2019.

    To help accelerate 3D deep learning research, released Kaolin, a library that provides efficient implementations of differentiable 3D modules for use in DL systems.

    63 proslijeđena tweeta 155 korisnika označava da im se sviđa
    Poništi
  9. proslijedio/la je Tweet
    5. stu 2019.

    ICLR papers with perfect scores (all 8s, total 11 papers): 1. "FreeLB: Enhanced Adversarial Training for Language Understanding" 2. "BackPACK: Packing more into Backprop"

    7 replies 191 proslijeđeni tweet 643 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    8. lis 2019.

    What space should diverse semantics be grounded & what should be the structure? 3D Scene Graph is a 4-layer structure for unified semantics, 3D space &camera. We demonstrate it on Gibson models with an automated labeling method. Data available to download!

    5 proslijeđenih tweetova 16 korisnika označava da im se sviđa
    Poništi
  11. proslijedio/la je Tweet
    1. lis 2019.

    DenseRaC: Joint 3D Pose and Shape Estimation by Dense Render-and-Compare.

    1 reply 27 proslijeđenih tweetova 102 korisnika označavaju da im se sviđa
    Poništi
  12. 28. kol 2019.

    Our ICCV 2019 paper on text-to-image matching is now on arxiv.

    1 proslijeđeni tweet 15 korisnika označava da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    26. kol 2019.

    I have a system to plan writing papers for conference deadlines. My students and some collaborators know about it. With the ICLR 2020 deadline coming up, I thought this might be a good time to share this with a wider audience.

    13 replies 245 proslijeđenih tweetova 982 korisnika označavaju da im se sviđa
    Poništi
  14. proslijedio/la je Tweet
    6. kol 2019.

    A Discussion of 'Adversarial Examples Are Not Bugs, They Are Features' - Six comments from the community and responses from the original authors.

    6 replies 218 proslijeđenih tweetova 673 korisnika označavaju da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  15. 25. lip 2019.

    I'm excited to share that I have joined Oculus Research in Sausalito, CA to work on 3D humans.

    1 reply 16 korisnika označava da im se sviđa
    Poništi
  16. proslijedio/la je Tweet
    14. lip 2019.

    Really excited to release Bayesian Deep Learning Benchmarks - please share with others who you think might like this, and have a look at the blog/repo/colab: This work was done over a period of a year and a half by many collaborators

    7 replies 396 proslijeđenih tweetova 1.143 korisnika označavaju da im se sviđa
    Poništi
  17. proslijedio/la je Tweet
    6. lip 2019.

    I'm excited to share our new paper that jointly detects objects and predicts 3D triangle meshes in real-world images, called Mesh R-CNN. With Georgia Gkioxari and Jitendra Malik

    8 replies 146 proslijeđenih tweetova 483 korisnika označavaju da im se sviđa
    Poništi
  18. proslijedio/la je Tweet
    23. svi 2019.

    Getting closer to the dream! A network that uses unlabelled images to boost performance when labels are scarce (new SOTA), and it's no worse than ResNet when labels are plentiful. Also: Unsupervised net + just a linear on top outperforms original AlexNet!

    5 replies 115 proslijeđenih tweetova 405 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    14. svi 2019.

    Pixel-aligned Implicit Function (PIFu), a new memory efficient, fully-convolutional 3D representation for recovering a fully textured surface of a clothed person from a single or multi-view image! With Shunsuke S, Zeng H, , Shigeo M,

    1 reply 117 proslijeđenih tweetova 415 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  20. proslijedio/la je Tweet
    25. tra 2019.

    New blog post: "A Recipe for Training Neural Networks" a collection of attempted advice for training neural nets with a focus on how to structure that process over time

    81 reply 1.908 proslijeđenih tweetova 5.364 korisnika označavaju da im se sviđa
    Poništi

@sarafianosn još nije objavio nijedan Tweet.

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:

    ·