Have you ever wondered how topological data analysis can be applied to the analysis of neural networks? I'm excited to share our newest survey on TDA for deep learning, offering an in-depth look at this intriguing question. 🧵
https://t.co/bzfkftB37f
📢The ChaLearn Looking at People #WACV2024 Challenge on Pedestrian Attribute Recognition and Attributed-based Person Retrieval is open on Codabench. Join us to push the boundaries of pedestrian attribute recognition along with concept drift.
Website: https://t.co/gDbGkCOXyj
Codalab open source platform already hosted thousands of competitions! Check its JMLR paper with public code: https://t.co/zihPSamxEh
With @Flowtus_Bot , I. Guyon, A. Letournel, D. Tran, Xavier Baro, @hugojair, T. Thomas, @NehzUx
Public instance: https://t.co/ZFq0dHCZ82
🔊 Check out 𝗕𝗲𝗟𝗙𝘂𝘀𝗶𝗼𝗻, our work now accepted at #ICCV2023, with my amazing supervisors @SergioEscalera_@cpalmeroc. In short:
𝗕𝗲havioral 𝗟atent Space ➕ Latent Dif𝗙𝘂𝘀𝗶𝗼𝗻
🟰 Realistic, coherent & diverse Human Motion Prediction (HMP)
More in the🧵 below!
[1/7]
CFP: Biometrics Security & Privacy, IJCV https://t.co/4J1ATjtqNG
Techniques producing realistic synthetic images proofs the need for better biometric systems. This SI focuses on detecting & defending against biometric attacks.
30th July, J.Wan A.Ross @iPRoBeLab P.Torr @OxfordTVG
Happy to announce our paper got accepted at #CVPR2023!
We generate cinemegraphs from images by animating clothes using a CycleNet trained without real data.
This is the result of my internship at #AdobeResearch.
Website: https://t.co/NZMS0WfQEb
arXiv: https://t.co/Ty2ZLXjAK3
Do you want to deal with the latest developments in deep learning for 3D human behavior understanding? Apply and join our HuPBA research team in Barcelona:
https://t.co/MbD71530Xp
Recently I had the opportunity to present in SIGGRAPH Asia 2022 my work in deep-based garment animation, Neural Cloth Simulation.
Our main contribution is unsupervised learning of cloth dynamics.
Visit the project site to learn more!
https://t.co/vHsXiTFcUa
I presented at SIGGRAPH Asia our work in cloth animation with deep learning.
-Unsupervised
-Physically based
-Multiple layers of cloth
-Extreme efficiency: +14K fps
-Simple formulation
https://t.co/sREDA3J4Y5
Video:
https://t.co/3VeVwn2eVT
Code:
https://t.co/et9k8eHDeM
The present and the future on Sign Language Recognition this Friday at CVPR 2021:
ChaLearn Looking at People Sign Language Recognition in the Wild Workshop (https://t.co/s0n8AzNSkW) and Challenge results (https://t.co/G2PyL0GIdt)
Thx to authors, competitors and invited speakers!
Happy to be by the third consecutive year among the top 3 winner teams at the EPIC KITCHENS action recognition challenge at CVPR. Happy to collaborate with this amazing team :) Thx a lot to the challenge organizers. @CVPR@dimadamen