@YSiglidis Profile picture

Yannis Siglidis

@YSiglidis

PhD student in Computer Vision; conceptual artist; tortured-philosopher; ex-poet

Joined November 2021
Similar User
Antoine Guédon photo

@antoine_guedon

Elliot Vincent photo

@Elt_Vincent

Lucas Ventura photo

@Lucas__Ventura

Imagine-ENPC photo

@ImagineEnpc

Monika Wysoczańska photo

@mkwysoczanska

Loic Landrieu photo

@loiclandrieu

Johannes Lutzeyer photo

@JLutzeyer

Katherine Tieu photo

@kthrn_uiuc

Gilles Puy photo

@gillespuy

Malamatenia Vlachou Efstathiou photo

@malamatenio_

Robin Courant photo

@robin_courant

Bruno Neri photo

@neribr

Yasser Benigmim photo

@yasserbenigmim

Thibault GROUEIX photo

@thibaultgroueix

M.Bülent Sarıyıldız photo

@mbsariyildiz

Pinned

The Learnable Typewriter A Generative Approach to Text Analysis w/ @NicaoGr @jgaubil @t_monnier and Mathieu Aubry ⭐️ won the best paper award at #ICDAR24 learnable-typewriter.github.io

Tweet Image 1

✮ Diffusion Models as Data Mining Tools ⛏️ at Milan #ECCV2024! w/ @holynski_, Alyosha Efros, Mathieu Aubry, @shiryginosar Join us! ⋆。° ✮ -------------- Poster: Oct 1st (10:30 | #107) Presentation: Sep 30th (15:30 | Suite 4) sites.google.com/cs.stanford.ed… 🔗diff-mining.github.io

Tweet Image 1

On Saturday @malamatenio_ gave a great talk on how the Learnable Typewriter can be used to enable interpretable palaeographic morphological script type analysis in #IWCP. Today we present the Learnable Typewriter itself during the 11:10 oral session in Parthenon room at #ICDAR.

Tweet Image 1

An Interpretable Deep Learning Approach for Morphological Script Type Analysis (#IWCP 2024) w/@YSiglidis @Himanis6 and Mathieu Aubry 📜🔡📊 learnable-handwriter.github.io Join us for our talk in Athens // session II Parthenon I (11:15-13:15) csmc.uni-hamburg.de/iwcp2024.html

Tweet Image 1


Last year, I left my Parisian chat noir, “infinity” ∞, at BAIR, as a post-covid ESA. I recently found out it was used to test OOD generalization… As robotics progresses, I decided to leave fidget slugs this time. Could also relieve progress in attention economy’s metabolic ADD.


Yannis Siglidis Reposted

Can GPTs generate infinite and diverse data for robotics? Introducing RoboGen, a generative robotic agent that keeps proposing new tasks, creating corresponding environments and acquiring novel skills autonomously! code: github.com/Genesis-Embodi… 👇🧵 (better with audio)


Labels made datasets and datasets trained diffusion models and diffusion models found some pixels to be more typical of these labels. Mining them hopefully lets us learn more about our visual datasets (and labels) than what we already knew. #ECCV2024 Thanks @_akhaliq + reposts❤️

Diffusion Models as Data Mining Tools discuss: huggingface.co/papers/2408.02… This paper demonstrates how to use generative models trained for image synthesis as tools for visual data mining. Our insight is that since contemporary generative models learn an accurate representation of…

Tweet Image 1


3 years ago I had an idea🪸generative models (then meaning gpt-2) could reconcile distant-reading and close-reading, a term I then coined as latent-reading. 3 years later this seedling grew it's own life and through the care of my collaborators became a CV paper🦋#ECCV2024

Image synthesis models can be used for visual data mining! See our new #ECCV2024 paper: "Diffusion Models as Data Mining Tools." Project page: diff-mining.github.io Paper: arxiv.org/abs/2408.02752 1/9

Tweet Image 1


Loading...

Something went wrong.


Something went wrong.