Ignacio Rocco

Ignacio Rocco

Senior Research Scientist

Google DeepMind

About

I am a Senior Research Scientist at Google DeepMind working on problems related to computer vision.

Previously, I was a post-doctoral researcher at Facebook AI Research (FAIR), working with Natalia Neverova and Andrea Vedaldi. I completed my PhD at the Willow team, INRIA/ENS, under the supervision of Josef Sivic and Relja Aranđelović, developing trainable methods for solving image alignment problems. Earlier, I earned a Master's degree in Mathematics/Vision/Machine Learning from ENS Paris-Saclay.

News

06/2025 TAPNext accepted at ICCV 2025.
05/2025 Direct Motion Models accepted at ICML 2025.
07/2024 CoTracker accepted at ECCV 2024.
09/2024 Tapvid-3D accepted at NeurIPS 2024.
05/2023 I started as a Research Scientist at Google DeepMind.
03/2023 2 accepted papers at CVPR 2023.
04/2022 Awarded the best PhD thesis prize by AFRIF.
03/2022 2 accepted papers at CVPR 2022.
04/2021 Started a post-doc at FAIR.
10/2020 Graduated from PhD. Find the manuscript here.

Publications

Efficiently Reconstructing Dynamic Scenes one D4RT at a Time
C. Zhang, G. Le Moing, S. Koppula, I. Rocco, L. Momeni, J. Xie, S. Sun, R. Sukthankar, J. K. Barral, R. Hadsell, Z. Ghahramani, A. Zisserman, J. Zhang, M. S. M. Sajjadi
arXiv 2025
Project Page arXiv Blog Post
Trajan
Direct Motion Models for Assessing Generated Videos
K. Allen, C. Doersch, G. Zhou, M. Suhail, D. Driess, I. Rocco, Y. Rubanova, T. Kipf, M. S. M. Sajjadi, K. Murphy, J. Carreira, S. van Steenkiste
ICML 2025
Project Page arXiv
TAPNext: Tracking Any Point (TAP) as Next Token Prediction
A. Zholus, C. Doersch, Y. Yang, S. Koppula, V. Pătrăucean, X. O. He, I. Rocco, M. S. M. Sajjadi, S. Chandar, R. Goroshin
ICCV 2025
Project Page arXiv
Scaling 4D representations
Scaling 4D Representations
J. Carreira, D. Gokay, M. King, C. Zhang, I. Rocco, A. Mahendran, T.A. Keck, J. Heyward, S. Koppula, E. Pot, G. Erdogan, Y. Hasson, Y. Yang, K. Greff, G. Le Moing, S. van Steenkiste, D. Zoran, D. A. Hudson, P. Vélez, L. Polanía, L. Friedman, C. Duvarney, R. Goroshin, K. Allen, J. Walker, R. Kabra, E. Aboussouan, J. Sun, T. Kipf, C. Doersch, V. Pătrăucean, D. Damen, P. Luc, M. S. M. Sajjadi, A. Zisserman
arXiv 2024
arXiv
TAPVid-3d
TAPVid-3d: A benchmark for tracking any point in 3d
S. Koppula, I. Rocco, Y. Yang, J. Heyward, J. Carreira, A. Zisserman, G. Brostow, C. Doersch
NeurIPS 2024
Project Page arXiv
BootsTAP
BootsTAP: Bootstrapped training for Tracking-Any-Point
C. Doersch, P. Luc, Y. Yang, D. Gokay, S. Koppula, A. Gupta, J. Heyward, I. Rocco, R. Goroshin, J. Carreira, A. Zisserman
ACCV 2024
Project Page arXiv
CoTracker: It is better to track together
N. Karaev, I. Rocco, B. Graham, N. Neverova, A. Vedaldi, C. Rupprecht
ECCV 2024
Project Page arXiv
Replay
Replay: Multi-modal Multi-view Acted Videos for Casual Holography
R. Shapovalov*, Y. Kleiman*, I. Rocco*, D. Novotny, A. Vedaldi, C. Chen, F. Kokkinos, B. Graham, N. Neverova
ICCV 2023
Project Page arXiv
Real-time volumetric rendering of dynamic humans
I. Rocco, I. Makarov, F. Kokkinos, D. Novotny, B. Graham, N. Neverova, A. Vedaldi
arXiv 2023
Project Page arXiv
DynamicStereo: Consistent Dynamic Depth from Stereo Videos
N. Karaev, I. Rocco, B. Graham, N. Neverova, A. Vedaldi, C. Rupprecht
CVPR 2023
Project Page arXiv
COP3D
Common pets in 3d: Dynamic new-view synthesis of real-life deformable categories
S. Sinha, R. Shapovalov, J. Reizenstein, I. Rocco, N. Neverova, A. Vedaldi, D. Novotny
CVPR 2023
Project Page arXiv
SyncMatch
Self-supervised correspondence estimation via multiview registration
M. El Banani, I. Rocco, D. Novotny, A. Vedaldi, N. Neverova, J. Johnson, B. Graham
WACV 2023
Project Page arXiv
KeyTr
KeyTr: Keypoint Transporter for 3D Reconstruction of Deformable Objects in Videos
D. Novotny, I. Rocco, S. Sinha, A. Carlier, G. Kerchenbaum, R. Shapovalov, N. Smetanin, N. Neverova, B. Graham, A. Vedaldi
CVPR 2022 (Oral)
Paper
BodyMap
BodyMap: Learning Full-Body Dense Correspondence Map
A. Ianina, N. Sarafianos, Y. Xu, I. Rocco, T. Tung
CVPR 2022
Project Page arXiv
Sparse NCNet
Efficient Neighbourhood Consensus Networks via Submanifold Sparse Convolutions
I. Rocco, R. Arandjelović and J. Sivic
ECCV 2020
Project Page Code
D2-Net
D2-Net: A Trainable CNN for Joint Detection and Description of Local Features
M. Dusmanu, I. Rocco, T. Pajdla, M. Pollefeys, J. Sivic, A. Torii, T. Sattler
CVPR 2019
Project Page Code
NCNet
Neighbourhood Consensus Networks
I. Rocco, M. Cimpoi, R. Arandjelović, A. Torii, T. Pajdla and J. Sivic
NeurIPS 2018 (Spotlight)
Project Page Code
CNN Geometric
Convolutional neural network architecture for geometric matching
I. Rocco, R. Arandjelović and J. Sivic
CVPR 2017 (Spotlight)
Project Page PyTorch

Teaching & Reviewing

Teaching assistant for the Object Recognition and Computer Vision course of ENS Paris Saclay's MVA M2 Master (2016, 2017, 2018).

I have reviewed for CVPR, ECCV, BMVC, NeurIPS, T-PAMI, and other journals.