Pith. sign in

REVIEW 4 major objections 6 minor 30 references

AnonyNoise: Anonymizing Event Data with Smart Noise to Outsmart Re-Identification and Preserve Privacy

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read AnonyNoise is a trainable noise-injection pipeline that anonymizes event-camera data against neural-network re-identification while keeping downstream tasks like gesture and emotion recognition usable.

desk verdict Real engineering contribution undone by an overclaimed privacy guarantee; worth reviewing with a mandatory language and robustness revision. read the letter →

arxiv 2411.16440 v1 pith:SHSTRLHT submitted 2024-11-25 cs.CV

Katharina Bendig , René Schuster , Nicole Thiemer , Karen Joisten , Didier Stricker This is my paper · ORCID
classification cs.CV
keywords eventcamerasprivacypreservationre-identificationanonymizationadversarialtraininghistogramsnoiseinjectiongesturerecognition
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that event-camera data, once assumed private because it is hard for humans to read, can still leak identity through neural networks, and that a learned noise injection can close that leak without destroying the data's usefulness. AnonyNoise adds data-dependent noise to event histograms, trained adversarially against a continuously adapting re-identification network. On two real event datasets the paper measures re-identification accuracy falling by roughly 60% and 50% relative to raw data, while target-task accuracy drops by about 12% and 22%. The anonymizer also transfers to an unseen dataset and resists denoising/inversion and grayscale-reconstruction attacks. A sympathetic reader would take this as evidence that practical, task-preserving event anonymization is possible, and that machine re-identification should be treated as a first-class privacy threat.

What carries the argument

The load-bearing object is the anonymization network $f_{\text{Anon}}$, a lightweight fully convolutional network with no skip connections that outputs per-pixel noise parameters. Its input is the event histogram $X_e$ concatenated with random noise $n \sim \mathcal{N}(0,1)$; its output is the anonymized histogram $X'_e = X_e + X_\mu + X_{\sigma^2} \cdot n$, a reparameterization that keeps the operation differentiable. Two auxiliary networks pull in opposite directions during training: a target network $f_T$ encourages preserving task-relevant structure, while a re-identification network $f_{\text{ReId}}$, adversarially updated, tries to recover identity; the anonymizer's loss is $L_{\text{Anon}} = L_T - \mathcal{T}(\hat{X}_{\text{id}})$, i.e., target cross-entropy minus the triplet loss on identity features. Post-training, $f_{\text{Anon}}$ is frozen and a simulated attacker retrains $f_{\text{ReId}}$ on the anonymized data, which is the evaluation that grounds the privacy claim.

What would settle it

Train a re-identification model with a substantially different architecture (for example, a transformer or a graph neural network operating directly on raw event streams) on the anonymized outputs of AnonyNoise, or scale up the denoising/inversion network; if re-identification accuracy on the anonymized data rises back to near the raw-data level, the central privacy claim is refuted.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a per-pixel, data-dependent noise generator can strip identity information from event streams while preserving the information a downstream task needs. The anonymized events are produced by $X'_e = X_e + X_\mu + X_{\sigma^2} \cdot n$, where the network predicts the noise statistics from the input histogram and the random draw. A min-max adversarial loop keeps the re-identification network adapting, so the anonymizer cannot settle for fooling one fixed attacker. In the post-training attack simulation, AnonyNoise lowers re-identification accuracy on DVS-Gesture from 98.33% to 40.00% and on SEE from 72.5% to 25.0%, while holding target accuracy at 80.68% and 55.71%; on unseen Event-ReId data, top-1 re-identification drops from 59.09% to 38.64% without fine-tuning. The paper further claims robustness to a learned denoising/inversion attack and to grayscale image reconstruction.

Load-bearing premise

The privacy claim rests on the assumption that the re-identification attacks tested (a single residual-network architecture and a single event-histogram representation) are representative of all realistic attackers; if a different architecture or a raw sparse-event representation lets an attacker recover identity from AnonyNoise output, the core privacy guarantee fails.

Editorial extensions

If this is right

  • If the central claim holds, event-based surveillance and always-on cameras can be deployed with a privacy layer that blocks automatic person tracking without sacrificing the sensing task the camera was installed for.
  • The same pipeline can be retargeted to any downstream task that should not depend on identity, such as emotion detection, activity recognition, or gaze estimation, by swapping the target network.
  • The generalization result on Event-ReId suggests that an anonymizer trained on one event domain can be transferred to new camera setups, resolutions, and body poses without retraining.
  • The robustness to inversion and reconstruction attacks means that adding a learned denoiser before classification does not restore the attacker's ability to identify individuals from AnonyNoise output.
  • Because the anonymization noise is data-dependent, fixed denoising filters cannot undo it, which is the qualitative advantage over simple Gaussian noise reported by the paper.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Inference: the reparameterized noise mechanism is generic and could be carried over to other sensor streams such as radar or LiDAR, where deep networks are beginning to enable re-identification.
  • Inference: since the anonymizer is trained in tandem with a specific target task, its privacy/utility balance may not transfer to a different downstream task; a task-agnostic or multi-task variant would be a natural next step.
  • Inference: the paper reports empirical robustness but no formal privacy guarantee; quantifying residual identity leakage with an information-theoretic or differential-privacy bound would turn the claimed protection into a provable one.
  • Inference: the attacker in the main experiments shares the same representation (event histograms) and a similar backbone as the target network; a harder attacker using raw sparse events or a very different architecture would be a sterner test of the re-identification claim.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper proposes AnonyNoise, a learned noise-prediction network that anonymizes event-camera histograms by adding data-dependent noise, with the stated goal of preventing re-identification by neural networks while preserving downstream-task accuracy. The pipeline trains an anonymization network adversarially against a ResNet50 re-identification network and a target network, then simulates a post-training attack by retraining the re-identification network on the frozen anonymized data. Experiments on DVS-Gesture, SEE, and Event-ReId report reductions in re-identification accuracy (up to about 60% on DVS-Gesture) with moderate target-task degradation, plus robustness to inversion and image-reconstruction attacks. The method is presented as the first event-data anonymization pipeline specifically targeting neural-network re-identification.

Significance. If the claims were fully supported, this would be a useful contribution: it addresses a timely privacy problem for event cameras, introduces an adversarial min-max training scheme with a post-training attack evaluation, and releases code. The paper is also honest about several limitations (e.g., histogram-only input). However, the central headline claim—that AnonyNoise "prevents" re-identification by neural networks—is contradicted by the paper's own numbers: post-anonymization re-identification accuracy remains far above chance on the same architecture used in training. The real contribution is a privacy-utility tradeoff method that reduces, rather than prevents, re-identification. Because the paper frames the contribution as prevention and claims robustness against arbitrary re-identification networks, the current evidence base is not sufficient to support the stated claims.

major comments (4)
  1. [Abstract, Sec. 7, Table 1] The paper's central claim that AnonyNoise "prevents re-identification ... by neural networks" is not supported by the reported results. In Table 1, the post-anonymization re-identification accuracy is 40.00% on DVS-Gesture and 25.0% on SEE. Even under the most favorable reading of the evaluation protocol, chance-level performance would be at most about 16.7% for DVS-Gesture (with 6 test subjects) and 5% for SEE (with 20 validation subjects); the reported numbers are therefore well above chance. This is not a speculation about stronger attackers: the same ResNet50 architecture used in adversarial training is the one that still re-identifies individuals after anonymization. The abstract's more measured phrasing, "reducing attackers' re-identification capabilities by up to 60%," is supported, but the stronger "prevent" claim appears throughout the abstract, introduction, and conclusion and should be revised or substantiated with near-chance residual accuracy.
  2. [Sec. 4.3, Eq. (4); Sec. 4.4; Contributions bullet] The claim that the adversarial training "ensures the robustness of our anonymization against exploitation from any re-identification network" is unjustified. The anonymization loss in Eq. (4) only penalizes the triplet term, not the classifier cross-entropy term of the attacker's loss in Eq. (2). More importantly, the post-training attack in Sec. 4.4 retrains the same ResNet50 backbone on the same event-histogram representation used during adversarial training. The residual re-identification accuracy of 40.00% on DVS-Gesture and 38.64% top-1 on Event-ReId (Table 2) demonstrates that even this same-architecture attacker retains substantial identity information. To support the robustness claim, the authors would need to evaluate attackers with different backbones, different event representations, or otherwise argue why the trained architecture is a worst-case proxy. Without this, the "any re-identification network" claim is unsupported.
  3. [Tables 1–3; Sec. 5.1–5.4] All quantitative results are reported as single runs with no error bars, confidence intervals, or multiple training seeds. This is particularly problematic for the privacy claims, because the re-identification validation sets are small: DVS-Gesture has only 6 test subjects, and SEE has 20 validation subjects with two query samples per subject. An accuracy value such as 25.0% on SEE corresponds to roughly 10 correct queries out of 40, and the difference between 25% and chance-level 5% may be sensitive to the random query/gallery split. Without repeated trials or a statistical significance measure, the paper cannot support precise statements such as "reducing accid by nearly 60%" or the comparison of ablation variants in Table 3.
  4. [Sec. 5.1, re-identification evaluation protocol] The construction of the DVS-Gesture re-identification evaluation is under-specified, and this materially affects the interpretation of "prevention." The paper states that DVS-Gesture has 6 test and 23 training subjects, and that a query set is built by choosing a random sample per subject and per target label, but it does not state whether the gallery/query sets are restricted to the test subjects or include training subjects. The baseline raw re-identification accuracy of 98.33% in Table 1 suggests the protocol may not be a standard closed-set test with only 6 identities. The authors should clarify the exact gallery and query construction, the number of query samples, and the resulting chance-level accuracy, since the residual risk claim depends directly on this protocol.
minor comments (6)
  1. [Introduction, Sec. 1] The sentence "latest research showed that neural networks are able to reconstruct high-quality grayscale images ... This proofs the remainder of structural and personal information" contains a typo: "proofs" should be "proves."
  2. [Sec. 3.1] The phrase "This poses a potential thread against the privacy of the individual" should read "threat" instead of "thread."
  3. [Sec. 4.4] The heading "Post-Training Attack" contains a typo: "re-identifaction" should be "re-identification."
  4. [Sec. 4.1] The notation for the event histogram is introduced as shape (2*T, H, W) with T=5, but the text later refers to "T = 5timesteps" and the equations do not explicitly use T. Please make the temporal-bin definition consistent and explicit in the equations.
  5. [Fig. 1] The caption states "The arrow is pointing from the result without to the result with a denoise network used," which is hard to parse. Please rephrase to clarify which points are compared.
  6. [Sec. 5.2] The sentence "As an optimizer we use AdamW [15]" is followed by a period but the preceding text in the same paragraph is not a complete sentence; please merge these into a grammatically correct paragraph.

Circularity Check

0 steps flagged · score 2.0 of 10

No significant circularity: the privacy claim is empirically tested on held-out data and an unseen dataset, with only a non-load-bearing self-citation.

full rationale

The derivation chain is self-contained. AnonyNoise is defined by Eq. (1) and trained with explicit losses (Eqs. 2-4); no output quantity is defined in terms of the quantity it is claimed to predict. The central privacy claim is tested empirically on held-out validation partitions (Table 1), on an unseen dataset without fine-tuning (Event-ReId, Table 2), and under post-training adaptive attacks including a denoising/inversion network (Sec. 5.6, Table 4). These evaluations could have failed, and in fact residual re-identification accuracy remains above chance, so the results are measurements rather than consequences of the training objective. The paper cites its own prior work [6] only for a data-augmentation choice (ShapeAug), which is not load-bearing. One evaluation limitation is that the adversarial trainer and the post-training attacker share the ResNet50 architecture and event-histogram representation, so the claim of robustness against 'any' re-identification network is stronger than what the experiments establish; this is a threat to external validity and a wording overclaim, but it is not circular because the reported accid values are not forced by construction.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The central result depends on the choice of event histogram representation, the attacker architecture used in adversarial training, and the specific query/gallery splits. These are not independently validated external benchmarks.

free parameters (3)
  • Input noise scale = N(0,1) for DVS-Gesture, N(0,0.1) for SEE
    The scale of the random noise concatenated with the event histograms is set per dataset and affects the strength of anonymization and downstream accuracy (Sec. 5.2, supplementary).
  • Training hyperparameters = See supplementary Table 5
    Batch size, learning rate, epochs, and schedulers are chosen separately for each dataset and training stage; these choices influence the reported accuracies.
  • Temporal bins T = 5
    Event streams are aggregated into 5 timesteps for both polarities; this resolution may affect the amount of identity information preserved in the histogram representation.
assumptions (3)
  • domain assumption Event histograms of 2*T channels are a sufficient representation for both re-identification and downstream tasks.
    The method operates only on histograms, not raw sparse events, so any identity information lost in aggregation is not measured (Sec. 4.1).
  • ad hoc to paper The ResNet50 backbone is a representative attacker model.
    The defense is trained against this specific architecture, and the post-training attack retrains it; robustness to other architectures is assumed, not tested (Sec. 4.2, 4.4).
  • domain assumption Randomly constructed query/gallery splits measure re-identification risk.
    The splits are generated by a random process and rely on small numbers of identities; the resulting accuracy estimates are noisy (Sec. 5.1).

how reviews work

0 comments
Cite this review

Pith. "Pith review of AnonyNoise: Anonymizing Event Data with Smart Noise to Outsmart Re-Identification and Preserve Privacy." pith.science (2026). https://pith.science/paper/SHSTRLHT

@misc{pith2026241116440,
  author       = {Pith},
  title        = {Pith review of: AnonyNoise: Anonymizing Event Data with Smart Noise to Outsmart Re-Identification and Preserve Privacy},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SHSTRLHT}},
  note         = {Machine review of arXiv:2411.16440}
}
read the original abstract

The increasing capabilities of deep neural networks for re-identification, combined with the rise in public surveillance in recent years, pose a substantial threat to individual privacy. Event cameras were initially considered as a promising solution since their output is sparse and therefore difficult for humans to interpret. However, recent advances in deep learning proof that neural networks are able to reconstruct high-quality grayscale images and re-identify individuals using data from event cameras. In our paper, we contribute a crucial ethical discussion on data privacy and present the first event anonymization pipeline to prevent re-identification not only by humans but also by neural networks. Our method effectively introduces learnable data-dependent noise to cover personally identifiable information in raw event data, reducing attackers' re-identification capabilities by up to 60%, while maintaining substantial information for the performing of downstream tasks. Moreover, our anonymization generalizes well on unseen data and is robust against image reconstruction and inversion attacks. Code: https://github.com/dfki-av/AnonyNoise

Figures

Figures reproduced from arXiv: 2411.16440 by the authors.

Figure 1
Figure 1. Visualization of re-identification and target accuracy for [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Our approach addresses all involved interests simultaneously. The target network (green) aims to optimize performance on the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Example visualizations of the raw and anonymized events for [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: For the inversion attack, we insert a denoising network [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Visualized examples of a) the original event data, b) the anonymized event, c) original events with Gaussian noise, d) grayscale [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

30 extracted references · 29 canonical work pages

  1. [1]

    Person re-identification without identification via event anonymiza- tion

    Shafiq Ahmad, Pietro Morerio, and Alessio Del Bue. Person re-identification without identification via event anonymiza- tion. In International Conference on Computer Vision (ICCV), 2023. 2, 3, 4, 5, 6, 8, 10, 11

    work page 2023

  2. [2]

    Event anonymization: Privacy-preserving person re-identification and pose estimation in event-based vision

    Shafiq Ahmad, Pietro Morerio, and Alessio Del Bue. Event anonymization: Privacy-preserving person re-identification and pose estimation in event-based vision. IEEE Access ,

    work page

  3. [3]

    Event-driven re-id: A new benchmark and method towards privacy-preserving person re-identification

    Shafiq Ahmad, Gianluca Scarpellini, Pietro Morerio, and Alessio Del Bue. Event-driven re-id: A new benchmark and method towards privacy-preserving person re-identification. In Winter Conference on Applications of Computer Vision Workshops (WACVW), 2022. 2, 10, 11, 12

    work page 2022

  4. [4]

    A low power, fully event-based gesture recognition system

    Arnon Amir, Brian Taba, David Berg, Timothy Melano, Jef- frey McKinstry, Carmelo Di Nolfo, Tapan Nayak, Alexander Andreopoulos, Guillaume Garreau, Marcela Mendoza, Jeff Kusnitz, Michael Debole, Steve Esser, Tobi Delbruck, My- ron Flickner, and Dharmendra Modha. A low power, fully event-based gesture recognition system. In Conference on Computer Vision and...

    work page 2017

  5. [5]

    Privatsph¨are 4.0

    Hauke Behrendt, Wulf Loh, Tobias Matzner, and Catrin Mis- selhorn. Privatsph¨are 4.0. Metzler, Stuttgart, 2019. 2

    work page 2019

  6. [6]

    Sha- peaug: Occlusion augmentation for event camera data

    Katharina Bendig, Ren ´e Schuster, and Didier Stricker. Sha- peaug: Occlusion augmentation for event camera data. InIn- ternational Conference on Pattern Recognition Applications and Methods (ICPRAM), 2023. 6

    work page 2023

  7. [7]

    Spact: Self-supervised privacy preservation for action recog- nition

    Ishan Rajendrakumar Dave, Chen Chen, and Mubarak Shah. Spact: Self-supervised privacy preservation for action recog- nition. In Conference on Computer Vision and Pattern Recognition (CVPR), 2022. 3

    work page 2022

  8. [8]

    Imagenet: A large-scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In Conference on Computer Vision and Pattern Recognition (CVPR), 2009. 4

    work page 2009

Show all 30 references
  1. [9]

    Ted-spad: Temporal distinctiveness for self- supervised privacy-preservation for video anomaly detec- tion

    Joseph Fioresi, Ishan Rajendrakumar Dave, and Mubarak Shah. Ted-spad: Temporal distinctiveness for self- supervised privacy-preservation for video anomaly detec- tion. In International Conference on Computer Vision (ICCV), 2023. 3

    2023

  2. [10]

    Sicherheits-und ¨uberwachungstechnik

    Sandro Gaycken. Sicherheits-und ¨uberwachungstechnik. Handbuch Technikethik, 2013. 2

    2013

  3. [11]

    Datasheets for datasets

    Timnit Gebru, Jamie Morgenstern, Briana Vecchione, Jen- nifer Wortman Vaughan, Hanna Wallach, Hal Daum´e Iii, and Kate Crawford. Datasheets for datasets. Communications of the ACM, 2021. 2

    2021

  4. [12]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Conference on Computer Vision and Pattern Recognition (CVPR), 2016. 4

    2016

  5. [13]

    Privacy- preserving action recognition via motion difference quan- tization

    Sudhakar Kumawat and Hajime Nagahara. Privacy- preserving action recognition via motion difference quan- tization. In European Conference on Computer Vision (ECCV), 2022. 3

    2022

  6. [14]

    Estimation of vehi- cle speed based on asynchronous data from a silicon retina optical sensor

    Martin Litzenberger, Bernhard Kohn, Ahmed Nabil Bel- bachir, Nikolaus Donath, Gerhard Gritsch, Heinrich Garn, Christoph Posch, and Stephan Schraml. Estimation of vehi- cle speed based on asynchronous data from a silicon retina optical sensor. In The International Conference on...

    2006

  7. [15]

    Decoupled weight de- cay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight de- cay regularization. In International Conference on Learning Representations (ICLR), 2017. 6, 10

    2017

  8. [16]

    Privacy in context: Technology, policy, and the integrity of social life

    Helen Nissenbaum. Privacy in context: Technology, policy, and the integrity of social life. In Privacy in Context. Stan- ford University Press, 2009. 2

    2009

  9. [17]

    Back to event basics: Self-supervised learning of image reconstruc- tion for event cameras via photometric constancy

    Federico Paredes-Vall ´es and Guido CHE de Croon. Back to event basics: Self-supervised learning of image reconstruc- tion for event cameras via photometric constancy. InConfer- ence on Computer Vision and Pattern Recognition (CVPR) ,

  10. [18]

    High speed and high dynamic range video with an event camera

    Henri Rebecq, Ren ´e Ranftl, Vladlen Koltun, and Davide Scaramuzza. High speed and high dynamic range video with an event camera. IIEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2019. 2, 3, 7, 8, 11, 12

    2019

  11. [19]

    Asynchronous event- based clustering and tracking for intrusion monitoring in uas

    Juan Pablo Rodr ´ıguez-G´omez, A G ´omez Egu ´ıluz, JR Mart´ınez-de Dios, and Anibal Ollero. Asynchronous event- based clustering and tracking for intrusion monitoring in uas. In International Conference on Robotics and Automa- tion (ICRA), 2020. 1, 2

    2020

  12. [20]

    Die Zukunft von Privatheit und Selbstbestimmung: Analysen und Empfehlungen zum Schutz der Grundrechte in der digitalen Welt

    Alexander Roßnagel and Michael Friedewald. Die Zukunft von Privatheit und Selbstbestimmung: Analysen und Empfehlungen zum Schutz der Grundrechte in der digitalen Welt. Springer Nature, 2022. 2

    2022

  13. [21]

    Privatsph ¨are

    Harald Seubert. Privatsph ¨are. In Handbuch Angewandte Ethik. Springer, 2023. 2

    2023

  14. [22]

    Eventmix: An efficient data augmentation strategy for event-based learning

    Guobin Shen, Dongcheng Zhao, and Yi Zeng. Eventmix: An efficient data augmentation strategy for event-based learning. Information Sciences, 2023. 6

    2023

  15. [23]

    Big Data und Privatheit, Novem- ber 2014

    Martin Steinebach, Oren Halvani, Marcel Sch ¨afer, Christian Winter, and York Yannikos. Big Data und Privatheit, Novem- ber 2014. Fraunhofer-Institut f ¨ur Sichere Informationstech- nologie SIT. 2

    2014

  16. [24]

    Iot-guard: Event-driven fog-based video surveillance system for real-time security management

    Tanin Sultana and Khan A Wahid. Iot-guard: Event-driven fog-based video surveillance system for real-time security management. IEEE Access, 2019. 1, 2

    2019

  17. [25]

    Event- based video reconstruction using transformer

    Wenming Weng, Yueyi Zhang, and Zhiwei Xiong. Event- based video reconstruction using transformer. In Interna- tional Conference on Computer Vision (ICCV), 2021. 2, 3

    2021

  18. [26]

    Die ethische Bewertung von Big Data

    Klaus Wiegerling. Die ethische Bewertung von Big Data . Springer Fachmedien Wiesbaden, 2023. 2

    2023

  19. [27]

    Privacy-preserving deep action recogni- tion: An adversarial learning framework and a new dataset

    Zhenyu Wu, Haotao Wang, Zhaowen Wang, Hailin Jin, and Zhangyang Wang. Privacy-preserving deep action recogni- tion: An adversarial learning framework and a new dataset. IEEE Transactions on Pattern Analysis and Machine Intelli- gence (TPAMI), 2020. 3

    2020

  20. [28]

    Deep learning for person re- identification: A survey and outlook

    Mang Ye, Jianbing Shen, Gaojie Lin, Tao Xiang, Ling Shao, and Steven CH Hoi. Deep learning for person re- identification: A survey and outlook. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) , 2021. 1, 3

    2021

  21. [29]

    In the blink of an eye: Event-based emotion recognition

    Haiwei Zhang, Jiqing Zhang, Bo Dong, Pieter Peers, Wen- wei Wu, Xiaopeng Wei, Felix Heide, and Xin Yang. In the blink of an eye: Event-based emotion recognition. In Inter- national Conference on Computer Graphics and Interactive Techniques (SIGGRAPH), 2023. 5, 6, 7, 10, 11, 12

    2023

  22. [30]

    Event-based video reconstruction via potential-assisted spiking neural network

    Lin Zhu, Xiao Wang, Yi Chang, Jianing Li, Tiejun Huang, and Yonghong Tian. Event-based video reconstruction via potential-assisted spiking neural network. In Conference on Computer Vision and Pattern Recognition (CVPR), 2022. 2, 3 Supplementary Material A. Overview In this sup...

    2022

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.