TY - GEN
T1 - MIGS
T2 - 18th European Conference on Computer Vision, ECCV 2024
AU - Chatziagapi, Aggelina
AU - Chrysos, Grigorios G.
AU - Samaras, Dimitris
N1 - Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.
PY - 2025
Y1 - 2025
N2 - We introduce MIGS (Multi-Identity Gaussian Splatting), a novel method that learns a single neural representation for multiple identities, using only monocular videos. Recent 3D Gaussian Splatting (3DGS) approaches for human avatars require per-identity optimization. However, learning a multi-identity representation presents advantages in robustly animating humans under arbitrary poses. We propose to construct a high-order tensor that combines all the learnable 3DGS parameters for all the training identities. By assuming a low-rank structure and factorizing the tensor, we model the complex rigid and non-rigid deformations of multiple subjects in a unified network, significantly reducing the total number of parameters. Our proposed approach leverages information from all the training identities and enables robust animation under challenging unseen poses, outperforming existing approaches. It can also be extended to learn unseen identities. Project page: https://aggelinacha.github.io/MIGS/.
AB - We introduce MIGS (Multi-Identity Gaussian Splatting), a novel method that learns a single neural representation for multiple identities, using only monocular videos. Recent 3D Gaussian Splatting (3DGS) approaches for human avatars require per-identity optimization. However, learning a multi-identity representation presents advantages in robustly animating humans under arbitrary poses. We propose to construct a high-order tensor that combines all the learnable 3DGS parameters for all the training identities. By assuming a low-rank structure and factorizing the tensor, we model the complex rigid and non-rigid deformations of multiple subjects in a unified network, significantly reducing the total number of parameters. Our proposed approach leverages information from all the training identities and enables robust animation under challenging unseen poses, outperforming existing approaches. It can also be extended to learn unseen identities. Project page: https://aggelinacha.github.io/MIGS/.
KW - 3D Gaussian Splatting
KW - Human Body Animation
KW - Multiple Identities
UR - https://www.scopus.com/pages/publications/85208587743
U2 - 10.1007/978-3-031-72691-0_22
DO - 10.1007/978-3-031-72691-0_22
M3 - Conference contribution
AN - SCOPUS:85208587743
SN - 9783031726903
T3 - Lecture Notes in Computer Science
SP - 388
EP - 408
BT - Computer Vision – ECCV 2024 - 18th European Conference, Proceedings
A2 - Leonardis, Aleš
A2 - Ricci, Elisa
A2 - Roth, Stefan
A2 - Russakovsky, Olga
A2 - Sattler, Torsten
A2 - Varol, Gül
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 29 September 2024 through 4 October 2024
ER -