---
_id: '18244'
abstract:
- lang: eng
text: Some face recognition methods are designed to utilize geometric information
extracted from depth sensors to overcome the weaknesses of single-image based
recognition technologies. However, the accurate acquisition of the depth profile
is an expensive and challenging process. Here, we introduce a novel method that
learns to recognize faces from stereo camera systems without the need to explicitly
compute the facial surface or depth map. The raw face stereo images along with
the location in the image from which the face is extracted allow the proposed
CNN to improve the recognition task while avoiding the need to explicitly handle
the geometric structure of the face. This way, we keep the simplicity and cost
efficiency of identity authentication from a single image, while enjoying the
benefits of geometric data without explicitly reconstructing it. We demonstrate
that the suggested method outperforms both existing single-image and explicit
depth based methods on largescale benchmarks, and even capable of recognize spoofing
attacks. We also provide an ablation study that shows that the suggested method
uses the face locations in the left and right images to encode informative features
that improve the overall performance.
article_number: '9320359'
article_processing_charge: No
arxiv: 1
author:
- first_name: Amir
full_name: Livne, Amir
last_name: Livne
- first_name: Ziv
full_name: Aviv, Ziv
last_name: Aviv
- first_name: Shahaf
full_name: Grofit, Shahaf
last_name: Grofit
- first_name: Alexander
full_name: Bronstein, Alexander
id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
last_name: Bronstein
orcid: 0000-0001-9699-8730
- first_name: Ron
full_name: Kimmel, Ron
last_name: Kimmel
citation:
ama: 'Livne A, Aviv Z, Grofit S, Bronstein AM, Kimmel R. Do we need depth in state-uf-the-art
face authentication? In: 2020 International Conference on 3D Vision (3DV).
IEEE; 2021. doi:10.1109/3dv50981.2020.00099'
apa: 'Livne, A., Aviv, Z., Grofit, S., Bronstein, A. M., & Kimmel, R. (2021).
Do we need depth in state-uf-the-art face authentication? In 2020 International
Conference on 3D Vision (3DV). Fukuoka, Japan: IEEE. https://doi.org/10.1109/3dv50981.2020.00099'
chicago: Livne, Amir, Ziv Aviv, Shahaf Grofit, Alex M. Bronstein, and Ron Kimmel.
“Do We Need Depth in State-Uf-the-Art Face Authentication?” In 2020 International
Conference on 3D Vision (3DV). IEEE, 2021. https://doi.org/10.1109/3dv50981.2020.00099.
ieee: A. Livne, Z. Aviv, S. Grofit, A. M. Bronstein, and R. Kimmel, “Do we need
depth in state-uf-the-art face authentication?,” in 2020 International Conference
on 3D Vision (3DV), Fukuoka, Japan, 2021.
ista: Livne A, Aviv Z, Grofit S, Bronstein AM, Kimmel R. 2021. Do we need depth
in state-uf-the-art face authentication? 2020 International Conference on 3D Vision
(3DV). 8th International Conference on 3D Vision, 9320359.
mla: Livne, Amir, et al. “Do We Need Depth in State-Uf-the-Art Face Authentication?”
2020 International Conference on 3D Vision (3DV), 9320359, IEEE, 2021,
doi:10.1109/3dv50981.2020.00099.
short: A. Livne, Z. Aviv, S. Grofit, A.M. Bronstein, R. Kimmel, in:, 2020 International
Conference on 3D Vision (3DV), IEEE, 2021.
conference:
end_date: 2020-11-28
location: Fukuoka, Japan
name: 8th International Conference on 3D Vision
start_date: 2020-11-25
date_created: 2024-10-08T13:04:02Z
date_published: 2021-01-19T00:00:00Z
date_updated: 2024-12-12T10:10:29Z
day: '19'
doi: 10.1109/3dv50981.2020.00099
extern: '1'
external_id:
arxiv:
- '2003.10895'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.2003.10895
month: '01'
oa: 1
oa_version: Preprint
publication: 2020 International Conference on 3D Vision (3DV)
publication_identifier:
eissn:
- 2475-7888
isbn:
- '9781728181295'
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: Do we need depth in state-uf-the-art face authentication?
type: conference
user_id: 3E5EF7F0-F248-11E8-B48F-1D18A9856A87
year: '2021'
...
---
_id: '18257'
abstract:
- lang: eng
text: We consider the problem of localizing relevant subsets of non-rigid geometric
shapes given only a partial 3D query as the input. Such problems arise in several
challenging tasks in 3D vision and graphics, including partial shape similarity,
retrieval, and non-rigid correspondence. We phrase the problem as one of alignment
between short sequences of eigenvalues of basic differential operators, which
are constructed upon a scalar function defined on the 3D surfaces. Our method
therefore seeks for a scalar function that entails this alignment. Differently
from existing approaches, we do not require solving for a correspondence between
the query and the target, therefore greatly simplifying the optimization process;
our core technique is also descriptor-free, as it is driven by the geometry of
the two objects as encoded in their operator spectra. We further show that our
spectral alignment algorithm provides a remarkably simple alternative to the recent
shape-from-spectrum reconstruction approaches. For both applications, we demonstrate
improvement over the state-of-the-art either in terms of accuracy or computational
cost.
article_number: '8886146'
article_processing_charge: No
arxiv: 1
author:
- first_name: Arianna
full_name: Rampini, Arianna
last_name: Rampini
- first_name: Irene
full_name: Tallini, Irene
last_name: Tallini
- first_name: Maks
full_name: Ovsjanikov, Maks
last_name: Ovsjanikov
- first_name: Alexander
full_name: Bronstein, Alexander
id: 58f3726e-7cba-11ef-ad8b-e6e8cb3904e6
last_name: Bronstein
orcid: 0000-0001-9699-8730
- first_name: Emanuele
full_name: Rodola, Emanuele
last_name: Rodola
citation:
ama: 'Rampini A, Tallini I, Ovsjanikov M, Bronstein AM, Rodola E. Correspondence-free
region localization for partial shape similarity via Hamiltonian spectrum alignment.
In: 2019 International Conference on 3D Vision (3DV). IEEE; 2019. doi:10.1109/3dv.2019.00014'
apa: 'Rampini, A., Tallini, I., Ovsjanikov, M., Bronstein, A. M., & Rodola,
E. (2019). Correspondence-free region localization for partial shape similarity
via Hamiltonian spectrum alignment. In 2019 International Conference on 3D
Vision (3DV). Quebec City, QC, Canada: IEEE. https://doi.org/10.1109/3dv.2019.00014'
chicago: Rampini, Arianna, Irene Tallini, Maks Ovsjanikov, Alex M. Bronstein, and
Emanuele Rodola. “Correspondence-Free Region Localization for Partial Shape Similarity
via Hamiltonian Spectrum Alignment.” In 2019 International Conference on 3D
Vision (3DV). IEEE, 2019. https://doi.org/10.1109/3dv.2019.00014.
ieee: A. Rampini, I. Tallini, M. Ovsjanikov, A. M. Bronstein, and E. Rodola, “Correspondence-free
region localization for partial shape similarity via Hamiltonian spectrum alignment,”
in 2019 International Conference on 3D Vision (3DV), Quebec City, QC, Canada,
2019.
ista: Rampini A, Tallini I, Ovsjanikov M, Bronstein AM, Rodola E. 2019. Correspondence-free
region localization for partial shape similarity via Hamiltonian spectrum alignment.
2019 International Conference on 3D Vision (3DV). 7th International Conference
on 3D Vision, 8886146.
mla: Rampini, Arianna, et al. “Correspondence-Free Region Localization for Partial
Shape Similarity via Hamiltonian Spectrum Alignment.” 2019 International Conference
on 3D Vision (3DV), 8886146, IEEE, 2019, doi:10.1109/3dv.2019.00014.
short: A. Rampini, I. Tallini, M. Ovsjanikov, A.M. Bronstein, E. Rodola, in:, 2019
International Conference on 3D Vision (3DV), IEEE, 2019.
conference:
end_date: 2019-09-19
location: Quebec City, QC, Canada
name: 7th International Conference on 3D Vision
start_date: 2019-09-16
date_created: 2024-10-08T13:07:51Z
date_published: 2019-10-31T00:00:00Z
date_updated: 2024-12-05T15:49:06Z
day: '31'
doi: 10.1109/3dv.2019.00014
extern: '1'
external_id:
arxiv:
- '1906.06226'
language:
- iso: eng
main_file_link:
- open_access: '1'
url: https://doi.org/10.48550/arXiv.1906.06226
month: '10'
oa: 1
oa_version: Preprint
publication: 2019 International Conference on 3D Vision (3DV)
publication_identifier:
eissn:
- 2475-7888
isbn:
- '9781728131320'
publication_status: published
publisher: IEEE
quality_controlled: '1'
scopus_import: '1'
status: public
title: Correspondence-free region localization for partial shape similarity via Hamiltonian
spectrum alignment
type: conference
user_id: 3E5EF7F0-F248-11E8-B48F-1D18A9856A87
year: '2019'
...