pith. sign in

arxiv: 2606.22378 · v1 · pith:MWBAOSS4new · submitted 2026-06-21 · 💻 cs.CV

Following the Flow: Advection-Consistent Modeling for Event-based Small Object Detection

Wen Guo , Fulong Cai , Wuzhou Quan This is my paper

Pith reviewed 2026-06-26 10:44 UTC · model grok-4.3

classification 💻 cs.CV
keywords event camerassmall object detectionadvection modelingfeature transportvelocity fieldstemporal consistencyphysics-guided modeling
0
0 comments X

The pith

Advection constraints along velocity fields preserve weak event responses for better small object detection.

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

The paper introduces PACT, a framework that models event camera data evolution as a motion-driven transport process rather than simple local aggregation. It propagates features using estimated velocity fields and applies advection constraints to keep trajectory consistency. This helps maintain the signals from small objects that would otherwise fragment due to sparsity and noise. Sympathetic readers would care because event cameras are promising for high-speed vision but struggle with tiny moving objects in real scenes. If effective, it makes reliable detection possible without losing temporal continuity.

Core claim

PACT integrates motion-aware feature extraction with a differentiable advection-based transport operator. This enforces coherent motion representation by propagating features along velocity fields and enforcing trajectory-level consistency through advection constraints, thereby preserving weak event responses over time.

What carries the argument

the motion-driven feature transport process using a differentiable advection-based transport operator that propagates features along estimated velocity fields

If this is right

  • Small object signals maintain temporal continuity instead of becoming fragmented.
  • Weak responses are protected from degradation by complex background interference.
  • Detection achieves 20.72% higher IoU and 15.03% higher accuracy on event-based benchmarks.
  • Computational efficiency remains comparable to prior methods while improving coherence.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar advection ideas could apply to other asynchronous sensor fusion tasks where motion estimation is feasible.
  • Longer event sequences might benefit more as consistency accumulates over extended trajectories.
  • The reliance on velocity accuracy suggests pairing with robust optical flow methods for events could further strengthen results.

Load-bearing premise

Estimated velocity fields from sparse events are accurate enough to act as reliable transport paths without adding significant errors to the features.

What would settle it

Running the model on synthetic event data where velocity fields are intentionally perturbed or estimated poorly and measuring if the IoU gains disappear.

Figures

Figures reproduced from arXiv: 2606.22378 by Fulong Cai, Wen Guo, Wuzhou Quan.

Figure 1
Figure 1. Figure 1: Overall architecture of PACT. The encoder stacks four T-FE modules. ATC estimates velocity fields and consistency scores at each scale. The decoder uses four A-FR modules with skip fusion to reconstruct temporally continuous trajectories. 2.2 Physics-guided Dynamics To stabilize motion cues in low-SNR event streams, many works introduce phys￾ical priors. A classic line is contrast maximization, which warps… view at source ↗
Figure 2
Figure 2. Figure 2: Architectures of the proposed modules. (a) T-FE filters and enhances sparse responses under trajectory-consistency guidance. (b) A-FR propagates features along the estimated velocity field to reconstruct continuous trajectories while suppressing incoherent triggers. its transported prediction. For each active voxel i with feature Fi , we predict a bounded 2D displacement ∆pi on the spatial grid: \Delta \ma… view at source ↗
Figure 3
Figure 3. Figure 3: Qualitative results of different methods. Correct detections are marked in green, false detections in red, missed detections in blue, and noise points in gray [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Advection residual density across training stages. Kernel density estimates show the advection residual Radv for target events and spurious triggers, with the corresponding detection metrics reported next to each stage. 4.4 Advection-Guided Representation Evolution PACT uses advection consistency to preserve weak, fragmented temporal re￾sponses while suppressing triggers that do not support coherent transp… view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of temporal continuity statistics across methods. We report the average length of continuous successful detection and the maximum length of contin￾uous detection loss per target [PITH_FULL_IMAGE:figures/full_fig_p012_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Comparison of temporal continuity visualizations across methods. We align each target’s detection trajectory to a unified time scale and visualize the hit ratio at each time step with color, where dark blue indicates no hit and yellow indicates a higher hit ratio. The right-side color strip groups targets by scene, with each block corresponding to one scenario, which makes it easy to compare continuous det… view at source ↗
Figure 7
Figure 7. Figure 7: Mechanism study of challenging scenes. (a) Three-frame visualization of a rep￾resentative case. Marker shape indicates time (circle: t, triangle: t+1, square: t+2), and color indicates target ID. Co-occurring targets exhibit markedly different displacements within the same window. (b) Scene-level statistics after removing simple scenes with￾out temporal target co-occurrence, since they do not form competin… view at source ↗
Figure 6
Figure 6. Figure 6: The target-index-∼40 group (Scene 9) exhibits higher Per-frame Target Density and the highest Per-frame Speed Outlierness. This indicates that more targets compete within the same window and that a few targets frequently move much faster than the rest, which makes a single first-order field insufficient to align all trajectories consistently. The statistics agree with the reduced margin in [PITH_FULL_IMAG… view at source ↗
Figure 1
Figure 1. Figure 1: Advection residual distributions under stochastic event dropping [PITH_FULL_IMAGE:figures/full_fig_p022_1.png] view at source ↗
read the original abstract

Event cameras enable high-frequency visual perception with microsecond latency, offering advantages for dynamic scenes. However, event-based small object detection remains challenging due to sparse asynchronous measurements and weak object responses that are easily disrupted by noise. Limited spatial support causes small-object signals to lose temporal continuity, resulting in fragmented and unstable predictions. To address this issue, we propose a physics-guided advection-consistent modeling framework, termed PACT, which formulates event evolution as a motion-driven feature transport process. Instead of relying solely on local spatio-temporal aggregation, PACT propagates features along estimated velocity fields and enforces trajectory-level consistency through advection constraints. This design preserves weak event responses over time and prevents their degradation under complex background interference. Technically, PACT integrates motion-aware feature extraction with a differentiable advection-based transport operator, enabling coherent motion representation and effective noise suppression during temporal evolution. Extensive experiments on benchmark event-based datasets demonstrate that PACT consistently outperforms state-of-the-art methods, achieving improvements of 20.72\% in IoU and 15.03\% in accuracy while maintaining comparable computational efficiency. The code is publicly available at https://github.com/fulongcai/PACT.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper proposes PACT, a physics-guided advection-consistent modeling framework for event-based small object detection. It formulates event evolution as a motion-driven feature transport process that propagates features along estimated velocity fields and enforces trajectory-level consistency via advection constraints, integrating motion-aware feature extraction with a differentiable advection-based transport operator. This is claimed to preserve weak event responses and suppress noise. Experiments on benchmark datasets report gains of 20.72% in IoU and 15.03% in accuracy over SOTA methods, with public code release.

Significance. If the central mechanism holds, the work offers a principled way to maintain temporal coherence for sparse, weak signals in event-based small-object detection, potentially improving robustness in dynamic scenes. Public code availability supports reproducibility and is a clear strength.

major comments (2)
  1. [Abstract] Abstract: the central claim that estimated velocity fields provide reliable transport paths for the differentiable advection operator is load-bearing, yet the abstract supplies no error analysis, sensitivity study, or ablation isolating velocity estimation accuracy; systematic misalignment from sparse/noisy events (as flagged in the stress-test note) could propagate rather than suppress degradation.
  2. [Abstract] Abstract: performance claims (20.72% IoU, 15.03% accuracy) are stated without reference to specific baselines, dataset statistics, or controls that would demonstrate the advection constraints are the source of the gains rather than ancillary design choices.
minor comments (1)
  1. Abstract lacks explicit dataset names and comparison methods despite claiming 'extensive experiments on benchmark event-based datasets'.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on the abstract. We agree that additional context on velocity estimation robustness and baseline specificity would strengthen the abstract and will revise accordingly while preserving its brevity.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim that estimated velocity fields provide reliable transport paths for the differentiable advection operator is load-bearing, yet the abstract supplies no error analysis, sensitivity study, or ablation isolating velocity estimation accuracy; systematic misalignment from sparse/noisy events (as flagged in the stress-test note) could propagate rather than suppress degradation.

    Authors: We acknowledge the abstract's brevity limits inclusion of detailed error analysis. The manuscript contains ablations on velocity estimation accuracy and sensitivity to noise in the experiments section, showing that the advection constraints mitigate misalignment effects rather than propagate them. We will revise the abstract to briefly note the robustness of the transport paths under sparse events and reference the supporting analyses. revision: yes

  2. Referee: [Abstract] Abstract: performance claims (20.72% IoU, 15.03% accuracy) are stated without reference to specific baselines, dataset statistics, or controls that would demonstrate the advection constraints are the source of the gains rather than ancillary design choices.

    Authors: The abstract states gains over state-of-the-art methods but does not name baselines due to length constraints. The full manuscript specifies comparisons on standard event-based datasets with ablations isolating the advection components. We will revise the abstract to reference the primary baselines and note that controls confirm the source of improvements. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation is self-contained

full rationale

The abstract and description present PACT as a physics-guided framework that integrates motion-aware extraction with a differentiable advection operator and enforces consistency via advection constraints. No equations, fitted parameters, or self-citations are shown that would reduce any claimed prediction or consistency result to a definition or input by construction. The central mechanism is described as an independent modeling choice rather than a renaming or self-referential fit. This matches the default case of a non-circular paper.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Only abstract available; ledger is therefore minimal and provisional.

axioms (1)
  • domain assumption Event evolution can be formulated as a motion-driven feature transport process
    Stated as the modeling choice that replaces local spatio-temporal aggregation.

pith-pipeline@v0.9.1-grok · 5733 in / 1126 out tokens · 30273 ms · 2026-06-26T10:44:27.189622+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

56 extracted references · 4 canonical work pages · 2 internal anchors

  1. [1]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    An, Z., Sun, G., Liu, Y., Liu, F., Wu, Z., Wang, D., Van Gool, L., Belongie, S.: Rethinking few-shot 3d point cloud semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 3996– 4006 (2024) Advection Consistency for Weak Event Responses 15

    2024

  2. [2]

    IEEE Transactions on Neural Networks and Learning Systems25(2), 407–417 (2014)

    Benosman, R., Clercq, C., Lagorce, X., Ieng, S.H., Bartolozzi, C.: Event-based visual flow. IEEE Transactions on Neural Networks and Learning Systems25(2), 407–417 (2014)

    2014

  3. [3]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Bi, Y., Chadha, A., Abbas, A., Bourtsoulatze, E., Andreopoulos, Y.: Graph- based object classification for neuromorphic vision sensing. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 491–501 (2019)

    2019

  4. [5]

    In: Proceedings of the International Joint Conference on Neural Networks

    Cordone, L., Miramond, B., Thierion, P.: Object detection with spiking neural networks on automotive event data. In: Proceedings of the International Joint Conference on Neural Networks. pp. 1–8. IEEE (2022)

    2022

  5. [6]

    IEEE Transactions on Pattern Analysis and Machine In- telligence44(1), 154–180 (2022)

    Gallego, G., Delbruck, T., Orchard, G., Bartolozzi, C., Taba, B., Censi, A., Leutenegger, S., Davison, A.J., Conradt, J., Daniilidis, K., Scaramuzza, D.: Event- based vision: A survey. IEEE Transactions on Pattern Analysis and Machine In- telligence44(1), 154–180 (2022)

    2022

  6. [7]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition

    Gallego, G., Rebecq, H., Scaramuzza, D.: A unifying contrast maximization frame- work for event cameras, with applications to motion, depth, and optical flow esti- mation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. pp. 3867–3876 (2018)

    2018

  7. [8]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Gehrig, D., Loquercio, A., Derpanis, K.G., Scaramuzza, D.: End-to-end learn- ing of representations for asynchronous event-based data. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 5632–5642 (2019)

    2019

  8. [9]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Gehrig, M., Scaramuzza, D.: Recurrent vision transformers for object detection with event cameras. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 13884–13893 (2023)

    2023

  9. [10]

    In: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion

    Hu, Q., Yang, B., Xie, L., Rosa, S., Guo, Y., Wang, Z., Trigoni, N., Markham, A.: Randla-net: Efficient semantic segmentation of large-scale point clouds. In: Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion. pp. 11108–11117 (2020)

    2020

  10. [11]

    In: Proceedings of the European Conference on Computer Vision

    Kim, H., Leutenegger, S., Davison, A.J.: Real-time 3d reconstruction and 6-dof tracking with an event camera. In: Proceedings of the European Conference on Computer Vision. Lecture Notes in Computer Science, vol. 9910, pp. 349–364. Springer (2016)

    2016

  11. [12]

    Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization (2014), arXiv:1412.6980

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  12. [13]

    IEEE Transactions on Pattern Analysis and Machine Intelligence39(7), 1346–1359 (2017)

    Lagorce, X., Orchard, G., Galluppi, F., Shi, B.E., Benosman, R.B.: Hots: A hier- archy of event-based time-surfaces for pattern recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence39(7), 1346–1359 (2017)

    2017

  13. [14]

    IEEE Journal of Solid-State Circuits 43(2), 566–576 (2008)

    Lichtsteiner, P., Posch, C., Delbruck, T.: A 128×128 120 db 15µs latency asyn- chronous temporal contrast vision sensor. IEEE Journal of Solid-State Circuits 43(2), 566–576 (2008)

    2008

  14. [15]

    In: Proceedings of the European Conference on Computer Vision

    Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., Berg, A.C.: Ssd: Single shot multibox detector. In: Proceedings of the European Conference on Computer Vision. pp. 21–37. Springer (2016)

    2016

  15. [16]

    In: Proceedings of the European Conference on Computer Vision

    Luo, X., Yao, M., Chou, Y., Xu, B., Li, G.: Integer-valued training and spike-driven inference spiking neural network for high-performance and energy-efficient object detection. In: Proceedings of the European Conference on Computer Vision. pp. 253–272. Springer (2024)

    2024

  16. [17]

    In: 16 W

    Maqueda, A.I., Loquercio, A., Gallego, G., García, N., Scaramuzza, D.: Event- based vision meets deep learning on steering prediction for self-driving cars. In: 16 W. Guo et al. ProceedingsoftheIEEE/CVFConferenceonComputerVisionandPatternRecog- nition. pp. 5419–5427 (2018)

    2018

  17. [18]

    In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems

    Mitrokhin, A., Fermüller, C., Parameshwara, C., Aloimonos, Y.: Event-based mov- ing object detection and tracking. In: Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems. pp. 1–9 (2018)

    2018

  18. [19]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Mitrokhin, A., Hua, Z., Fermüller, C., Aloimonos, Y.: Learning visual motion seg- mentation using event surfaces. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 14414–14423 (2020)

    2020

  19. [20]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Mondal, A., Giraldo, J.H., Bouwmans, T., Chowdhury, A.S.: Moving object de- tection for event-based vision using graph spectral clustering. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 876–884. IEEE (2021)

    2021

  20. [21]

    IEEE Signal Processing Magazine36(6), 51–63 (2019)

    Neftci, E.O., Mostafa, H., Zenke, F.: Surrogate gradient learning in spiking neural networks: Bringing the power of gradient-based optimization to spiking neural networks. IEEE Signal Processing Magazine36(6), 51–63 (2019)

    2019

  21. [22]

    IEEE Transactions on Pattern Analysis and Machine Intelligence44(5), 2519–2533 (2022)

    Pan, L., Hartley, R., Scheerlinck, C., Liu, M., Yu, X., Dai, Y.: High frame rate video reconstruction based on an event camera. IEEE Transactions on Pattern Analysis and Machine Intelligence44(5), 2519–2533 (2022)

    2022

  22. [23]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Peng, Y., Li, H., Zhang, Y., Sun, X., Wu, F.: Scene-adaptive sparse transformer for event-based object detection. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 16794–16804 (2024)

    2024

  23. [24]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Peng, Y., Zhang, Y., Xiong, Z., Sun, X., Wu, F.: Get: Group event transformer for event-based vision. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 6038–6048 (2023)

    2023

  24. [25]

    In: Advances in Neural Information Processing Systems

    Perot, E., de Tournemire, P., Nitti, D., Masci, J., Sironi, A.: Learning to detect objects with a 1 megapixel event camera. In: Advances in Neural Information Processing Systems. vol. 33, pp. 16639–16652 (2020)

    2020

  25. [26]

    IEEE Transactions on Pattern Analysis and Machine Intelligence43(6), 1964–1980 (2021)

    Rebecq, H., Ranftl, R., Koltun, V., Scaramuzza, D.: High speed and high dynamic range video with an event camera. IEEE Transactions on Pattern Analysis and Machine Intelligence43(6), 1964–1980 (2021)

    1964

  26. [27]

    Advances in Neural Information Processing Systems28(2015)

    Ren, S., He, K., Girshick, R., Sun, J.: Faster r-cnn: Towards real-time object de- tection with region proposal networks. Advances in Neural Information Processing Systems28(2015)

    2015

  27. [28]

    In: Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition

    Schaefer, S., Gehrig, D., Scaramuzza, D.: Aegnn: Asynchronous event-based graph neural networks. In: Proceedings of the IEEE/CVF Conference on Computer Vi- sion and Pattern Recognition. pp. 12371–12381. IEEE (2022)

    2022

  28. [29]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops

    Sekikawa, Y., Nagata, J.: Tangentially elongated gaussian belief propagation for event-based incremental optical flow estimation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. pp. 21940– 21949 (2023)

    2023

  29. [30]

    IEEE Access12, 51275–51306 (2024)

    Shariff, W., Dilmaghani, M.S., Kielty, P., Moustafa, M., Lemley, J., Corcoran, P.: Event cameras in automotive sensing: A review. IEEE Access12, 51275–51306 (2024)

    2024

  30. [31]

    In: Proceedings of the IEEE International Conference on Computer Vision

    Stoffregen, T., Gallego, G., Drummond, T., Kleeman, L., Scaramuzza, D.: Event- based motion segmentation by motion compensation. In: Proceedings of the IEEE International Conference on Computer Vision. pp. 7244–7253 (2019)

    2019

  31. [32]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Su, Q., Chou, Y., Hu, Y., Li, J., Mei, S., Zhang, Z., Li, G.: Deep directly-trained spiking neural networks for object detection. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 6555–6565 (2023) Advection Consistency for Weak Event Responses 17

    2023

  32. [33]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Thomas, H., Qi, C.R., Deschaud, J.E., Marcotegui, B., Goulette, F., Guibas, L.J.: Kpconv: Flexible and deformable convolution for point clouds. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 6411–6420 (2019)

    2019

  33. [34]

    Advances in Neural Information Processing Systems37, 107984–108011 (2024)

    Wang, A., Chen, H., Liu, L., Chen, K., Lin, Z., Han, J.: Yolov10: Real-time end- to-end object detection. Advances in Neural Information Processing Systems37, 107984–108011 (2024)

    2024

  34. [35]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Wang, X., Jin, Y., Wu, W., Zhang, W., Zhu, L., Jiang, B., Tian, Y.: Object de- tection using event camera: A moe heat conduction based detector and a new benchmark dataset. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 29321–29330 (2025)

    2025

  35. [36]

    In: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition

    Yang, C., Huang, Z., Wang, N.: Querydet: Cascaded sparse query for accelerating high-resolution small object detection. In: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition. pp. 13668–13677 (2022)

    2022

  36. [37]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

    Zhang,J.,Dong,B.,Zhang,H.,Ding,J.,Heide,F.,Yin,B.,Yang,X.:Spikingtrans- formers for event-based single object tracking. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 8801–8810. IEEE (2022)

    2022

  37. [38]

    In: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition

    Zhu, A.Z., Yuan, L., Chaney, K., Daniilidis, K.: Unsupervised event-based learning of optical flow, depth, and egomotion. In: Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition. pp. 989–997 (2019)

    2019

  38. [39]

    Deformable DETR: Deformable Transformers for End-to-End Object Detection

    Zhu, X., Su, W., Lu, L., Li, B., Wang, X., Dai, J.: Deformable detr: Deformable transformers for end-to-end object detection (2020), arXiv:2010.04159 Following the Flow: Advection-Consistent001 001 Modeling for Event-based Small Object Detection002 002 Supplementary Material003 003 Anonymous ECCV 2026 Submission004 004 Paper ID #13150005 005 1 Additional ...

    work page internal anchor Pith review Pith/arXiv arXiv 2020

  39. [40]

    IEEE Transactions on Neural Networks and Learning Systems25(2),126 126 407–417 (2014)127 127

    Benosman, R., Clercq, C., Lagorce, X., Ieng, S.H., Bartolozzi, C.: Event-based125 125 visual flow. IEEE Transactions on Neural Networks and Learning Systems25(2),126 126 407–417 (2014)127 127

    2014

  40. [41]

    Chen, N., Xiao, C., Dai, Y., He, S., Li, M., An, W.: Event-based tiny object128 128 detection: A benchmark dataset and baseline (2025), arXiv:2506.23575129 129

    work page arXiv 2025

  41. [42]

    de Tournemire, P., Nitti, D., Perot, E., Migliore, D., Sironi, A.: A large scale event-130 130 based detection dataset for automotive (2020), arXiv:2001.08499131 131

    work page arXiv 2020

  42. [43]

    In: Proceedings of the IEEE/CVF Conference on Computer133 133 Vision and Pattern Recognition

    Gallego, G., Gehrig, M., Scaramuzza, D.: Focus is all you need: Loss functions for132 132 event-based vision. In: Proceedings of the IEEE/CVF Conference on Computer133 133 Vision and Pattern Recognition. IEEE (2019)134 134

    2019

  43. [44]

    In: Proceedings of the IEEE Conference on Computer Vision and Pattern137 137 Recognition

    Gallego, G., Rebecq, H., Scaramuzza, D.: A unifying contrast maximization frame-135 135 work for event cameras, with applications to motion, depth, and optical flow esti-136 136 6 ECCV 2026 Submission #13150 mation. In: Proceedings of the IEEE Conference on Computer Vision and Pattern137 137 Recognition. pp. 3867–3876 (2018)138 138

    2026

  44. [45]

    In: Proceedings of the140 140 IEEE/CVF International Conference on Computer Vision

    Gehrig, D., Loquercio, A., Derpanis, K.G., Scaramuzza, D.: End-to-end learn-139 139 ing of representations for asynchronous event-based data. In: Proceedings of the140 140 IEEE/CVF International Conference on Computer Vision. pp. 5632–5642 (2019)141 141

    2019

  45. [46]

    In: Proceedings of the IEEE/CVF Conference on Computer143 143 Vision and Pattern Recognition

    Gehrig, M., Scaramuzza, D.: Recurrent vision transformers for object detection142 142 with event cameras. In: Proceedings of the IEEE/CVF Conference on Computer143 143 Vision and Pattern Recognition. pp. 13884–13893 (2023)144 144

    2023

  46. [47]

    In: Proceedings of the IEEE/CVF Conference on146 146 Computer Vision and Pattern Recognition

    Li, D., Li, J., Liu, X., Fan, X., Tian, Y.: Asynchronous collaborative graph repre-145 145 sentation for frames and events. In: Proceedings of the IEEE/CVF Conference on146 146 Computer Vision and Pattern Recognition. pp. 1655–1666. IEEE (2025)147 147

    2025

  47. [48]

    In: Proceedings of the European Confer-149 149 ence on Computer Vision

    Messikommer, N., Gehrig, D., Loquercio, A., Scaramuzza, D.: Event-based asyn-148 148 chronous sparse convolutional networks. In: Proceedings of the European Confer-149 149 ence on Computer Vision. pp. 415–431. Springer (2020)150 150

    2020

  48. [49]

    In: Proceedings of the IEEE/RSJ International152 152 Conference on Intelligent Robots and Systems

    Mitrokhin, A., Fermüller, C., Parameshwara, C., Aloimonos, Y.: Event-based mov-151 151 ing object detection and tracking. In: Proceedings of the IEEE/RSJ International152 152 Conference on Intelligent Robots and Systems. pp. 1–9 (2018)153 153

    2018

  49. [50]

    In: Advances in Neural Information155 155 Processing Systems

    Perot, E., de Tournemire, P., Nitti, D., Masci, J., Sironi, A.: Learning to detect154 154 objects with a 1 megapixel event camera. In: Advances in Neural Information155 155 Processing Systems. vol. 33, pp. 16639–16652 (2020)156 156

    2020

  50. [51]

    In: Proceedings of the IEEE/CVF Conference on Computer Vi-158 158 sion and Pattern Recognition

    Schaefer, S., Gehrig, D., Scaramuzza, D.: Aegnn: Asynchronous event-based graph157 157 neural networks. In: Proceedings of the IEEE/CVF Conference on Computer Vi-158 158 sion and Pattern Recognition. pp. 12371–12381. IEEE (2022)159 159

    2022

  51. [52]

    In: Proceedings of the IEEE/CVF International Conference on161 161 Computer Vision

    Shiba, S., Aoki, Y., Gallego, G.: Simultaneous motion and noise estimation with160 160 event cameras. In: Proceedings of the IEEE/CVF International Conference on161 161 Computer Vision. pp. 6959–6969. IEEE (2025)162 162

    2025

  52. [53]

    In: Proceedings of the IEEE164 164 International Conference on Computer Vision

    Stoffregen, T., Gallego, G., Drummond, T., Kleeman, L., Scaramuzza, D.: Event-163 163 based motion segmentation by motion compensation. In: Proceedings of the IEEE164 164 International Conference on Computer Vision. pp. 7244–7253 (2019)165 165

    2019

  53. [54]

    In: Proceedings of the IEEE/CVF167 167 International Conference on Computer Vision

    Torbunov, D., Ren, Y., Ghose, A., Dim, O., Cui, Y.: Evrt-detr: Latent space adap-166 166 tation of image detectors for event-based vision. In: Proceedings of the IEEE/CVF167 167 International Conference on Computer Vision. pp. 9812–9821. IEEE (2025)168 168

    2025

  54. [55]

    In: Proceedings of the IEEE/CVF170 170 Conference on Computer Vision and Pattern Recognition Workshops

    Yamaki, R., Shiba, S., Gallego, G., Aoki, Y.: Iterative event-based motion segmen-169 169 tation by variational contrast maximization. In: Proceedings of the IEEE/CVF170 170 Conference on Computer Vision and Pattern Recognition Workshops. pp. 4918–171 171

  55. [56]

    In: Proceedings174 174 of the IEEE/CVF International Conference on Computer Vision

    Zubić, N., Gehrig, D., Gehrig, M., Scaramuzza, D.: From chaos comes order: Or-173 173 dering event representations for object recognition and detection. In: Proceedings174 174 of the IEEE/CVF International Conference on Computer Vision. pp. 12846–12856.175 175 IEEE (2023)176 176

    2023

  56. [57]

    Zubic, N., Gehrig, M., Scaramuzza, D.: State space models for event cameras.177 177 In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern178 178 Recognition. pp. 5819–5828. IEEE (2024)179 179

    2024