Pith. sign in

REVIEW 3 major objections 5 minor 44 references

EvRT-DETR: Latent Space Adaptation of Image Detectors for Event-based Vision

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Frozen image detector plus tiny memory beats event-camera specialists.

desk verdict Solid two-stage recipe and a genuine Gen1 SOTA; the 1Mpx margin is within protocol noise. read the letter →

arxiv 2412.02890 v3 pith:HW4LRZUQ submitted 2024-12-03 cs.CV

Dmitrii Torbunov , Yihui Ren , Animesh Ghose , Odera Dim , Yonggang Cui This is my paper · ORCID
classification cs.CV
keywords event-basedvisionobjectdetectionRT-DETRlatentspaceadaptationConvLSTMtemporalmemoryGen1dataset1Mpx
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

Event-based cameras produce sparse asynchronous streams, and object detection on them has typically required bespoke representations and architectures. This paper claims that a mainstream image detector, RT-DETR, trained on nothing more than 50 ms stacks of event histograms, already matches specialized event-camera methods, and that adding small recurrent memory modules to the frozen detector's encoder features makes it the new state of the art on both standard automotive benchmarks, Gen1 (52.7 mAP, +2.3) and 1Mpx (50.1 mAP, +1.4). The framework, called I2EvDet, keeps the base detector frozen and trains only lightweight ConvLSTM adapters inserted at three feature scales, so the improvement comes with a few extra milliseconds of inference time. For a reader, the point is that event vision may not need custom architectures: a general recipe for turning any image detector into a temporal one could transfer to other asynchronous visual domains.

What carries the argument

The load-bearing component is the Temporal Adaptation Module: a ConvLSTM unit (a convolutional recurrent network) with down/up projection matrices, placed at each of the three encoder scales E3, E4, E5 of a frozen RT-DETR and combined with the spatial features through a residual addition with learnable scaling (ReZero-style). Together with the input representation—a Stacked 2D Histogram that packs events from a 50 ms window into a 20-channel image (ten 5 ms bins × two polarities)—this module is what lets the image detector ingest temporal context without retraining its spatial weights. The authors find the lowest feature scale contributes most, and a 4× reduction of the recurrent hidden dimension costs only 0.6 mAP while cutting trainable parameters to 2.3 M, a LoRA-like regime.

What would settle it

Find or construct a long stationary-sequence clip from Gen1 or 1Mpx where the vehicle stays still for more than ten seconds with dense traffic, and run EvRT-DETR: if its detections of cars and pedestrians drop to near-zero once the recurrent memory window (about 10 frames) expires, the paper's central claim that ConvLSTM memory maintains object persistence during event silence is wrong.

Watch

Extended reading notes

Core claim

The central discovery is that the latent representation space of a frozen image detector is a sufficient substrate for temporal event reasoning. After training RT-DETR on stacked 2D histograms of event data (20 channels: 2 polarities × 10 five-millisecond bins), the paper freezes its backbone and encoder and inserts three ConvLSTM modules, one per encoder feature scale, coupled through residual connections with learned scaling. These adapters give the model a memory that keeps detecting objects after they stop generating events—when the camera is stationary, event streams go quiet and frame-only detectors fail—and this accounts for the full gain, from 47.6 to 52.7 mAP on Gen1 and from 45.2 to 50.1 mAP on 1Mpx. The paper also reports that the same recipe improves YOLOX detectors by 4.4–6.4 mAP, indicating the mechanism is not specific to transformers.

Load-bearing premise

The entire result rests on the premise that a fixed 50 ms stacked histogram—ten 5 ms bins, resized and padded to RT-DETR's input shape—preserves enough of the event stream's information for automotive detection; the interpolation experiments on 1Mpx show that even the downsampling filter can move base mAP by 2.9 points, so if a richer temporal representation (time surfaces, voxel grids) were necessary for harder scenes, the reported gains would not transfer.

Editorial extensions

If this is right

  • If the claim holds, event-based object detection no longer needs specialized architectures: a frozen mainstream detector plus small recurrent adapters is enough to set new benchmarks.
  • The I2EvDet recipe generalizes across detector families: applying it to YOLOX yields consistent gains of 4.4–6.4 mAP on Gen1, not just to RT-DETR.
  • The temporal memory directly addresses the 'static object disappears when camera stops' failure mode of event cameras, as shown in the stop-and-go automotive sequences.
  • The framework is parameter-efficient in a LoRA-like way: using hidden dimension 64 (2.3 M trainable parameters) still beats all prior methods, so deployment cost is low.

Reading between the lines

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

  • If the simple 50 ms histogram is really enough, then even lighter detectors (mobile backbones, knowledge-distilled variants) should inherit the same +5 mAP gain from the adapters; that is a cheap experiment the paper did not run.
  • The 1Mpx result is sensitive to the downsampling interpolant (nearest vs bilinear changes base mAP by 2.9 points), suggesting the preprocessing pipeline, not the architecture, may set the ceiling; a testable extension is to learn the downsampling or representation jointly with the adapters.
  • The same frozen-detector-plus-recurrent-adapter recipe could transfer to other asynchronous sensors (tactile, audio, radar-like streams) wherever data arrive as sparse timed events rather than dense frames.
  • One could replace ConvLSTM with a state-space or attention-based memory to test whether recurrence is essential for object persistence or just sufficient.
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

3 major / 5 minor

Summary. The paper introduces I2EvDet, a two-stage framework for adapting frame-based object detectors to event-based camera data. In the first stage, an RT-DETR detector is trained on 20-channel stacked 2D histograms of 50 ms event windows. In the second stage, the detector is frozen and lightweight ConvLSTM modules are inserted at the three encoder scales, with residual fusion, to add temporal memory. The resulting EvRT-DETR models are evaluated on Gen1 and 1Mpx and are reported to reach 52.7 and 50.1 mAP, respectively, surpassing prior specialized EBC methods. The paper also reports ablations on module placement, hidden size, kernel size, augmentation, interpolation, clip length, and generalization to YOLOX.

Significance. If the headline results hold, the paper makes a useful contribution by showing that a mainstream detector can be adapted to event data with minimal architectural changes, and the code release and detailed ablations support reproducibility. The strongest aspects are the breadth of ablations, the YOLOX generalization experiments, and the parameter-efficiency analysis. The main caveat is that the 1Mpx SOTA margin is small relative to the paper's own measured sensitivities in preprocessing and hyperparameter choices, and no uncertainty estimates are provided.

major comments (3)
  1. [§4.1 and Supplementary E.2, Tables 1 and 11] The choice of 10-frame clips for 1Mpx was made by tuning on the same benchmark used for the headline claim; it yields +0.3 mAP (49.8 to 50.1), which is about 21% of the reported +1.4 margin. Please report results with a held-out validation split for hyperparameter selection, or provide multi-seed runs and confidence intervals, so that readers can assess whether the margin is distinguishable from protocol variation and training noise.
  2. [§4.1 and Supplementary E.1, Table 10] The interpolation method used to downsample 1Mpx changes the base RT-DETR-B mAP by 2.9 points (bilinear 45.2 vs. nearest 42.3 vs. bicubic 43.1), which is larger than the reported 1Mpx SOTA margin of +1.4. Since prior methods such as RVT, S5-ViT-B, ASTMNet, SAST-CB, and ERGO-12 use their own preprocessing pipelines, the 50.1 mAP result is directly comparable only if the exact frame construction and evaluation protocols are identical; please document the protocols of the prior comparisons or re-evaluate them under a common protocol.
  3. [Tables 1, 3, 4, 7-12] All results appear to be from single training runs without seeds or variance estimates. Given that the headline Gen1 margin is +2.3 and the 1Mpx margin is +1.4, please report standard deviation over at least three seeds for the principal comparisons, or otherwise justify that the reported differences exceed run-to-run variation.
minor comments (5)
  1. [Table 12] RT-DETR-B Gen1 mAP is listed as 47.5 in Table 12, whereas Tables 1, 3, and 7 report 47.6; please reconcile the inconsistency.
  2. [Equations (1)-(4)] The text says W_down is integrated within the RNN, but the equations present it as a separate projection; please align the notation with the actual implementation.
  3. [Table 10] The heading describes 'interpolation methods on base detector performance for the 1Mpx dataset downsampling'; please clarify that the comparison is for the downsample interpolation only and not for other preprocessing steps.
  4. [Figure 1] The marker for ERGO-12 reports a 100 ms runtime, but Table 1 shows a converted value; please state the conversion method in the caption so that the reader can interpret the time axis.
  5. [Supplement A.5] The end-to-end training experiments are mentioned but not quantified; either report the numbers or state explicitly that they are omitted for brevity.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the SOTA claims are empirical benchmark comparisons against external baselines, with no target metric defined by fitted constants or by self-citation.

full rationale

The paper's central claims are empirical: RT-DETR trained on stacked 2D event histograms reaches a competitive baseline, and adding frozen-backbone ConvLSTM temporal modules (EvRT-DETR) improves mAP on Gen1 and 1Mpx to 52.7 and 50.1, respectively. These numbers are evaluated on held-out test splits with the external Prophesee COCO-style toolbox, against baselines from other groups. No equation in the paper defines the reported mAP in terms of a fitted parameter, and no predicted quantity is algebraically equivalent to a training input or to the parameters that were tuned. The frame construction parameters (50 ms window, 10 bins) are adopted from RVT [12], but RVT is an external prior work by different authors; this is a protocol choice for comparability, not a self-citation or an ansatz smuggled from the authors' own prior work. The choice of ConvLSTM is likewise justified by reference to the external RVT model and to the original ConvLSTM paper. The two-stage training is a standard fit-then-freeze procedure, and the ablations over interpolation method, clip length, hidden size, kernel size, and module placement are performance-selection analyses, not circular reductions. The 1Mpx interpolation sensitivity (2.9 mAP difference between bilinear and nearest) and the 5-to-10 frame clip-length tuning are legitimate concerns about cross-method comparability and protocol sensitivity, but they do not make the derivation circular; they are external-validity and reproducibility risks. There are no load-bearing uniqueness theorems, no self-citational chains that force the result, and no renaming of a known empirical pattern as a new contribution. The paper is self-contained against external benchmarks, so the appropriate finding is no significant circularity.

Assumptions & free parameters 7 free parameters · 4 assumptions · 0 invented entities

No new physical entities or fitted theoretical constants appear. The claims rest on hand-selected architecture and preprocessing choices listed above, each validated only through single-run benchmark experiments.

free parameters (7)
  • ConvLSTM hidden dimension M = 256
    Ablation in Section 4.4.2: M=64 gives 52.1 mAP, M=256 gives 52.7, M=512 gives 52.9; M=256 is used in all reported models.
  • ConvLSTM kernel size = 3x3
    Ablation Table 8: KS=1 gives 52.3, KS=3 gives 52.7, KS=5 gives 52.0; KS=3 is adopted.
  • Temporal clip length for 1Mpx = 10 frames
    Supplement E.2 compares 5 vs 10 frames on 1Mpx (49.8 vs 50.1 mAP) and the 10-frame setting is used in the final SOTA run.
  • Event frame window and bins = 50 ms frame, 10 x 5 ms bins
    Adopted from RVT in Section 3.1, not derived; defines the entire input representation for all experiments.
  • 1Mpx downsampling target and interpolant = 360x640, bilinear
    Supplement E.1: nearest gives 42.3, bilinear 45.2, bicubic 43.1; bilinear is used in the final 1Mpx results.
  • Augmentation magnitudes and probabilities = flip 0.5, rotation 30 deg/0.6, translation 0.5/0.6, scale (0.5,1.5)/0.6, shear 30 deg/0.6, erase 0.4
    Hand-selected chain in Table 6; removing all augmentations drops RT-DETR-B from 47.6 to 38.6 mAP (Table 7).
  • Temporal module placement = scales E3, E4, E5 all on
    Ablation Table 3: (1,1,1) gives 52.7, while (0,1,1) gives 52.2, (1,0,1) gives 52.4, and (1,1,0) gives 51.0.
assumptions (4)
  • domain assumption Stacked 2D histograms over 50 ms preserve enough information for automotive object detection
    Section 3.1 converts events into a (20, H, W) frame and all downstream results depend on this representation; prior work [28] reported poor results with simple histograms, so this is a substantive assumption.
  • domain assumption COCO mAP with Prophesee's Automotive Dataset Toolbox is a comparable metric across all cited methods
    Section 4.1: SOTA comparisons rest on this metric; runtimes are further standardized by a FLOPS/throughput conversion, which is not a validated equivalence.
  • domain assumption An RNN hidden state can act as unbounded temporal memory to keep detecting static objects after events stop
    Section 3.3 and Figure 4: the temporal module is motivated by this claim; if the memory is insufficient for long stationary periods, the SOTA advantage over frame-based RT-DETR would shrink.
  • domain assumption Downsampling 1Mpx from 1280x720 to 640x360 does not remove detection-critical details
    Section 4.1 and Supplement E.1: all 1Mpx numbers use this resolution; the choice of interpolant changes base mAP by up to 2.9 points, so resolution reduction is consequential.

how reviews work

0 comments
Cite this review

Pith. "Pith review of EvRT-DETR: Latent Space Adaptation of Image Detectors for Event-based Vision." pith.science (2026). https://pith.science/paper/HW4LRZUQ

@misc{pith2026241202890,
  author       = {Pith},
  title        = {Pith review of: EvRT-DETR: Latent Space Adaptation of Image Detectors for Event-based Vision},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HW4LRZUQ}},
  note         = {Machine review of arXiv:2412.02890}
}
abstract

Event-based cameras (EBCs) have emerged as a bio-inspired alternative to traditional cameras, offering advantages in power efficiency, temporal resolution, and high dynamic range. However, the development of image analysis methods for EBCs is challenging due to the sparse and asynchronous nature of the data. This work addresses the problem of object detection for EBC cameras. The current approaches to EBC object detection focus on constructing complex data representations and rely on specialized architectures. We introduce I2EvDet (Image-to-Event Detection), a novel adaptation framework that bridges mainstream object detection with temporal event data processing. First, we demonstrate that a Real-Time DEtection TRansformer, or RT-DETR, a state-of-the-art natural image detector, trained on a simple image-like representation of the EBC data achieves performance comparable to specialized EBC methods. Next, as part of our framework, we develop an efficient adaptation technique that transforms image-based detectors into event-based detection models by modifying their frozen latent representation space through minimal architectural additions. The resulting EvRT-DETR model reaches state-of-the-art performance on the standard benchmark datasets Gen1 (mAP $+2.3$) and 1Mpx/Gen4 (mAP $+1.4$). These results demonstrate a fundamentally new approach to EBC object detection through principled adaptation of mainstream architectures, offering an efficient alternative with potential applications to other temporal visual domains. The code is available at: https://github.com/realtime-intelligence/evrt-detr

Figures

Figures reproduced from arXiv: 2412.02890 by the authors.

Figure 1
Figure 1. Object Detection Performance vs. Inference Time. A summary of object detection performance (COCO mAP) versus inference time (ms) of various models on the 1Mpx automotive dataset. Circle size is proportional to the number of model pa￾rameters. Inference times are reported for NVIDIA T4 GPU. Our EvRT-DETR models achieve state-of-the-art accuracy while main￾taining competitive inference speeds. ventional cameras that c… view at source ↗
Figure 2
Figure 2. Event-based Camera Output Visualization. Sample frames from the 1Mpx dataset showing event polarity. Red indi￾cates positive events, and blue denotes negative events. techniques. Early attempts to directly apply conventional image-based object detectors to simple event representa￾tions yielded poor performance [28], highlighting the gap between these domains. This led to two predominant re￾search directions: (1) con… view at source ↗
Figure 3
Figure 3. Overview of the I2EvDet Framework Applied to RT-DETR. This approach adapts a pre-trained RT-DETR model to process temporal event data using minimal architectural modifications. The backbone and encoder components remain frozen during temporal adaptation, while trainable temporal modules are strategically inserted at multiple feature scales. Left: Event camera frames are processed through the frozen RT-DETR backbone … view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Temporal Adaptation Benefits of EvRT-DETR. Visualization of object detection across sequential 1Mpx frames captured by a vehicle-mounted camera that gradually comes to a stop. While the camera is in motion (left frames), both detectors maintain accurate bounding boxes …
Figure 5
Figure 5. Figure 5: Qualitative detection results on diverse 1Mpx automotive scenarios. Top row: RT-DETR predictions. Middle row: EvRT￾DETR predictions. Bottom row: Ground truth annotations. Bounding box colors indicate object classes: red (cars), blue (pedestrians), orange (two-wheelers)…
Figure 6
Figure 6. Figure 6: Temporal sequence demonstrating event-based detection challenges during motion transitions. A vehicle approaches an intersection, stops (creating sparse event data), then resumes motion. Frames shown every 100 frames for visualization clarity. Top row: RT-DETR predicti…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

44 extracted references · 40 canonical work pages

  1. [1]

    Thomas Bachlechner, Bodhisattwa Prasad Majumder, Huanru Henry Mao, Gary Cottrell, and Julian J. McAuley. ReZero is all you need: fast convergence at large depth. In Proceedings of the Thirty-Seventh Conference on Uncer- tainty in Artificial Intelligence, UAI 2021, Virtual Event, 27- 30 July 2021, pages 1352–1361. AUAI Press, 2021. 6, 11

    work page 2021

  2. [2]

    End-to- end object detection with transformers

    Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, and Sergey Zagoruyko. End-to- end object detection with transformers. In Computer Vision - ECCV 2020 - 16th European Conference, Glasgow, UK, August 23-28, 2020, Proceedings, Part I , pages 213–229. Springer, 2020. 2, 3

    work page 2020

  3. [3]

    Ob- ject detection with spiking neural networks on automotive event data

    Lo ¨ıc Cordone, Benoˆıt Miramond, and Philippe Thi´erion. Ob- ject detection with spiking neural networks on automotive event data. In International Joint Conference on Neural Net- works, IJCNN 2022, Padua, Italy, July 18-23, 2022 , pages 1–8. IEEE, 2022. 3, 7

    work page 2022

  4. [4]

    Ekin Dogus Cubuk, Barret Zoph, Dandelion Man´e, Vijay Va- sudevan, and Quoc V . Le. AutoAugment: learning augmen- tation policies from data. CoRR, abs/1805.09501, 2018. 12

    arXiv 2018

  5. [5]

    Cubuk, Barret Zoph, Jonathon Shlens, and Quoc V

    Ekin D. Cubuk, Barret Zoph, Jonathon Shlens, and Quoc V . Le. Randaugment: Practical automated data augmentation with a reduced search space. In 2020 IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition, CVPR Workshops 2020, Seattle, WA, USA, June 14-19, 2020, pages 3008–3017. Computer Vision Foundation / IEEE, 2020. 12

    work page 2020

  6. [6]

    A large scale event-based detec- tion dataset for automotive

    Pierre de Tournemire, Davide Nitti, Etienne Perot, Davide Migliore, and Amos Sironi. A large scale event-based detec- tion dataset for automotive. CoRR, abs/2001.08499, 2020. 2, 6

    arXiv 2001

  7. [7]

    Qlora: Efficient finetuning of quantized llms

    Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer. Qlora: Efficient finetuning of quantized llms. In Advances in Neural Information Processing Systems 36: Annual Conference on Neural Information Processing Sys- tems 2023, NeurIPS 2023, New Orleans, LA, USA, Decem- ber 10 - 16, 2023, 2023. 4

    work page 2023

  8. [8]

    Event-based vision: A survey

    Guillermo Gallego, Tobi Delbr ¨uck, Garrick Orchard, Chiara Bartolozzi, Brian Taba, Andrea Censi, Stefan Leutenegger, Andrew J Davison, J ¨org Conradt, Kostas Daniilidis, et al. Event-based vision: A survey. IEEE transactions on pattern analysis and machine intelligence, 44(1):154–180, 2020. 1, 3, 4

    work page 2020

Show all 44 references
  1. [9]

    YOLOX: exceeding YOLO series in 2021

    Zheng Ge, Songtao Liu, Feng Wang, Zeming Li, and Jian Sun. YOLOX: exceeding YOLO series in 2021. CoRR, abs/2107.08430, 2021. 7, 11, 13

    2021 arXiv

  2. [10]

    YOLOX is a high-performance anchor-free YOLO, exceeding yolov3 v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported

    Zheng Ge, Songtao Liu, Feng Wang, Zeming Li, and Jian Sun. YOLOX is a high-performance anchor-free YOLO, exceeding yolov3 v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. https://github. com/Megvii-BaseDetection/YOLOX, 2024. 11

    2024

  3. [11]

    Pushing the limits of asynchronous graph-based object detection with event cam- eras

    Daniel Gehrig and Davide Scaramuzza. Pushing the limits of asynchronous graph-based object detection with event cam- eras. CoRR, abs/2211.12324, 2022. 3

    2022 arXiv

  4. [12]

    Recurrent vi- sion transformers for object detection with event cameras

    Mathias Gehrig and Davide Scaramuzza. Recurrent vi- sion transformers for object detection with event cameras. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023, Vancouver, BC, Canada, June 17- 24, 2023, pages 13884–13893. IEEE, 2023. 3, 4, 6, 7, 11, 13, 14

    2023

  5. [13]

    Girshick, Jeff Donahue, Trevor Darrell, and Jiten- dra Malik

    Ross B. Girshick, Jeff Donahue, Trevor Darrell, and Jiten- dra Malik. Rich feature hierarchies for accurate object de- tection and semantic segmentation. In 2014 IEEE Confer- ence on Computer Vision and Pattern Recognition, CVPR 2014, Columbus, OH, USA, June 23-28, 2014 , pages 580–

    2014

  6. [14]

    Parameter-efficient fine-tuning for large models: A comprehensive survey

    Zeyu Han, Chao Gao, Jinyang Liu, Jeff Zhang, and Sai Qian Zhang. Parameter-efficient fine-tuning for large models: A comprehensive survey. Trans. Mach. Learn. Res. , 2024,

    2024

  7. [15]

    Towards a unified view of parameter-efficient transfer learning

    Junxian He, Chunting Zhou, Xuezhe Ma, Taylor Berg- Kirkpatrick, and Graham Neubig. Towards a unified view of parameter-efficient transfer learning. In The Tenth Interna- tional Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022. OpenReview.net,...

    2022

  8. [16]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV , USA, June 27-30, 2016, pages 770–778. IEEE Computer Society, 2016. 3

    2016

  9. [17]

    Parameter-efficient transfer learning for NLP

    Neil Houlsby, Andrei Giurgiu, Stanislaw Jastrzebski, Bruna Morrone, Quentin de Laroussilhe, Andrea Gesmundo, Mona Attariyan, and Sylvain Gelly. Parameter-efficient transfer learning for NLP. In Proceedings of the 36th Interna- tional Conference on Machine Learning, ICML 2019, ...

    2019

  10. [18]

    Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen

    Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. LoRA: low-rank adaptation of large language models. In The Tenth International Conference on Learning Represen- tations, ICLR 2022, Virtual Event, April 25-29, 2022....

    2022

  11. [19]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. Adam: A method for stochastic optimization. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings, 2015. 11

    2015

  12. [20]

    Asynchronous spatio-temporal memory net- work for continuous event-based object detection

    Jianing Li, Jia Li, Lin Zhu, Xijie Xiang, Tiejun Huang, and Yonghong Tian. Asynchronous spatio-temporal memory net- work for continuous event-based object detection. IEEE Trans. Image Process., 31:2975–2987, 2022. 3, 7 9

    2022

  13. [21]

    Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll ´ar, and C

    Tsung-Yi Lin, Michael Maire, Serge J. Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll ´ar, and C. Lawrence Zitnick. Microsoft COCO: common objects in context. In Computer Vision - ECCV 2014 - 13th Eu- ropean Conference, Zurich, Switzerland, September 6-12, 2014, ...

    2014

  14. [22]

    Path aggregation network for instance segmentation

    Shu Liu, Lu Qi, Haifang Qin, Jianping Shi, and Jiaya Jia. Path aggregation network for instance segmentation. In2018 IEEE Conference on Computer Vision and Pattern Recog- nition, CVPR 2018, Salt Lake City, UT, USA, June 18- 22, 2018, pages 8759–8768. Computer Vision Foundation...

    2018

  15. [23]

    TorchVision: Py- Torch’s computer vision library

    TorchVision maintainers and contributors. TorchVision: Py- Torch’s computer vision library. https://github. com/pytorch/vision, 2016. 12

    2016

  16. [24]

    Event-based asynchronous sparse con- volutional networks

    Nico Messikommer, Daniel Gehrig, Antonio Loquercio, and Davide Scaramuzza. Event-based asynchronous sparse con- volutional networks. In Computer Vision - ECCV 2020 - 16th European Conference, Glasgow, UK, August 23-28, 2020, Proceedings, Part VIII, pages 415–431. Springer, 2020. 7

    2020

  17. [25]

    Better and faster: Adaptive event conversion for event-based object detection

    Yansong Peng, Yueyi Zhang, Peilin Xiao, Xiaoyan Sun, and Feng Wu. Better and faster: Adaptive event conversion for event-based object detection. In Thirty-Seventh AAAI Con- ference on Artificial Intelligence, AAAI 2023, Thirty-Fifth Conference on Innovative Applications of Art...

    2023

  18. [26]

    GET: group event transformer for event-based vision

    Yansong Peng, Yueyi Zhang, Zhiwei Xiong, Xiaoyan Sun, and Feng Wu. GET: group event transformer for event-based vision. In IEEE/CVF International Conference on Computer Vision, ICCV 2023, Paris, France, October 1-6, 2023, pages 6015–6025. IEEE, 2023. 2, 3, 7

    2023

  19. [27]

    Scene adaptive sparse transformer for event-based object detection

    Yansong Peng, Hebei Li, Yueyi Zhang, Xiaoyan Sun, and Feng Wu. Scene adaptive sparse transformer for event-based object detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, WA, USA, June 16-22, 2024, pages 16794–16804. IEEE, 2024. 3, 7

    2024

  20. [28]

    Learning to detect objects with a 1 megapixel event camera

    Etienne Perot, Pierre de Tournemire, Davide Nitti, Jonathan Masci, and Amos Sironi. Learning to detect objects with a 1 megapixel event camera. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Infor- mation Processing Systems 2020, NeurIPS 2...

    2020

  21. [29]

    Prophesee’s automotive dataset tool- box

    Prophesee. Prophesee’s automotive dataset tool- box. https : / / github . com / prophesee - ai/prophesee- automotive- dataset-toolbox,

  22. [30]

    Girshick, and Ali Farhadi

    Joseph Redmon, Santosh Kumar Divvala, Ross B. Girshick, and Ali Farhadi. You only look once: Unified, real-time ob- ject detection. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016, Las Vegas, NV , USA, June 27-30, 2016, pages 779–788. IEEE Compute...

    2016

  23. [31]

    Girshick, and Jian Sun

    Shaoqing Ren, Kaiming He, Ross B. Girshick, and Jian Sun. Faster R-CNN: towards real-time object detection with re- gion proposal networks. In Advances in Neural Information Processing Systems 28: Annual Conference on Neural In- formation Processing Systems 2015, December 7-12...

    2015

  24. [32]

    AEGNN: asynchronous event-based graph neural networks

    Simon Schaefer, Daniel Gehrig, and Davide Scaramuzza. AEGNN: asynchronous event-based graph neural networks. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022, New Orleans, LA, USA, June 18- 24, 2022, pages 12361–12371. IEEE, 2022. 3, 7

    2022

  25. [33]

    Convolutional LSTM network: A machine learning approach for precipitation nowcasting

    Xingjian Shi, Zhourong Chen, Hao Wang, Dit-Yan Yeung, Wai-Kin Wong, and Wang-chun Woo. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. In Advances in Neural Information Process- ing Systems 28: Annual Conference on Neural Information Proce...

    2015

  26. [34]

    Super-convergence: Very fast training of neural networks using large learn- ing rates

    Leslie N Smith and Nicholay Topin. Super-convergence: Very fast training of neural networks using large learn- ing rates. In Artificial intelligence and machine learning for multi-domain operations applications , pages 369–386. SPIE, 2019. 11

    2019

  27. [35]

    Gomez, Lukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Advances in Neu- ral Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017...

    2017

  28. [36]

    Ni, and Heung-Yeung Shum

    Hao Zhang, Feng Li, Shilong Liu, Lei Zhang, Hang Su, Jun Zhu, Lionel M. Ni, and Heung-Yeung Shum. DINO: DETR with improved denoising anchor boxes for end-to-end ob- ject detection. In The Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda,...

    2023

  29. [37]

    Official RT-DETR (RTDETR paddle pytorch), Real-Time DEtection TRansformer, DETRs beat YOLOs on Real-time object detection

    Yian Zhao, Wenyu Lv, Shangliang Xu, Jinman Wei, Guanzhong Wang, Qingqing Dang, Yi Liu, and Jie Chen. Official RT-DETR (RTDETR paddle pytorch), Real-Time DEtection TRansformer, DETRs beat YOLOs on Real-time object detection. https://github.com/lyuwenyu/ RT-DETR. 11

  30. [38]

    DETRs beat YOLOs on Real-time object detection

    Yian Zhao, Wenyu Lv, Shangliang Xu, Jinman Wei, Guanzhong Wang, Qingqing Dang, Yi Liu, and Jie Chen. DETRs beat YOLOs on Real-time object detection. In IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, WA, USA, June 16-22, 2024, pages 16965–16...

    2024

  31. [39]

    Convolution meets lora: Parameter efficient finetuning for segment anything model

    Zihan Zhong, Zhiqiang Tang, Tong He, Haoyang Fang, and Chun Yuan. Convolution meets lora: Parameter efficient finetuning for segment anything model. In The Twelfth In- ternational Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024 . OpenReview.net,

    2024

  32. [40]

    Deformable DETR: deformable transformers for end-to-end object detection

    Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, and Jifeng Dai. Deformable DETR: deformable transformers for end-to-end object detection. In 9th International Con- ference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net, 2021. 3 10

    2021

  33. [41]

    From chaos comes order: Ordering event representations for object recognition and detection

    Nikola Zubic, Daniel Gehrig, Mathias Gehrig, and Davide Scaramuzza. From chaos comes order: Ordering event representations for object recognition and detection. In IEEE/CVF International Conference on Computer Vision, ICCV 2023, Paris, France, October 1-6, 2023, pages 12800– 1...

    2023

  34. [42]

    State space models for event cameras

    Nikola Zubic, Mathias Gehrig, and Davide Scaramuzza. State space models for event cameras. In IEEE/CVF Con- ference on Computer Vision and Pattern Recognition, CVPR 2024, Seattle, WA, USA, June 16-22, 2024 , pages 5819–

    2024

  35. [587]

    IEEE Computer Society, 2014. 3

    2014

  36. [5828]

    3, 7, 13 A

    IEEE, 2024. 3, 7, 13 A. Training Setup This section details the implementation of the I2EvDet framework as applied to event-based object detection. The training methodology reflects a two-stage adaptation ap- proach: first establishing a robust spatial detector founda- tion th...

    2024

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.