Pith. sign in

REVIEW 4 major objections 6 minor 36 references

DEYOLO: Dual-Feature-Enhancement YOLO for Cross-Modality Object Detection

T0 review · 4 major / 6 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 This paper claims that DEYOLO, a YOLOv8-based detector, outperforms state-of-the-art single-modality and fusion-and-detection methods on RGB-infrared object detection in poor illumination by fusing modalities in feature space with…

desk verdict A sensible incremental RGB-IR detection architecture with a clean ablation, but the headline 'clear margin' claim is refuted by the paper's own Table 4. read the letter →

arxiv 2412.04931 v1 pith:W3NH7FYH submitted 2024-12-06 cs.CV

Yishuo Chen , Boran Wang , Xinyu Guo , Wenbin Zhu , Jiasheng He , Xiaobin Liu , Jing Yuan This is my paper · ORCID
classification cs.CV
keywords cross-modalityobjectdetectionRGB-infraredfusionfeature-leveldualenhancementchannelattentionspatiallow-lightYOLOv8
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

The paper tries to establish that fusing visible and infrared images inside a YOLO detector, rather than fusing the images first and then detecting, improves object detection in low-light scenes, and that its specific dual-enhancement modules reduce the mutual interference between the two modalities. If that is right, poor-illumination detection can be improved by a plug-in feature-space fusion that does not require a separate image-fusion stage. The claim is supported by experiments on the M3FD and LLVIP datasets, where the proposed DEYOLO reports higher mAP than single-modality detectors and fusion-and-detection pipelines.

What carries the argument

The machinery is a pair of dual-enhancement modules inserted between the YOLOv8 backbone and the detection heads. DECA computes channel weights: one from each modality and one from the concatenated mixed feature, multiplies them to enhance the mixed feature, then multiplies each original modality feature by the other modality's enhanced weight. DEPA does the same in the spatial dimension with convolutional pixel weights, and the enhanced RGB and IR features are added for detection. A third component, the bi-directional decoupled focus, splits the early feature map into two pixel groups for horizontal and vertical downsampling and concatenates them, enlarging the receptive field while preserving surrounding information. The dual-enhancement loop, where single modality enhances fusion and fusion enhances single modality, is what carries the claimed reduction of cross-modality interference.

What would settle it

A reader could settle it by taking the same 3:1:1 split, retraining each fusion-and-detection baseline and DEYOLO under identical schedules, and reading mAP50 on the test portion: if any baseline matches or beats 85.7 on M3FD or 96.4 on LLVIP, the paper's central margin claim fails.

Watch

Extended reading notes

Core claim

The central discovery claimed is that cross-modality RGB-infrared object detection benefits from a dual-enhancement fusion in feature space: a channel-weight module (DECA) and a spatial-weight module (DEPA) each first use single-modality information to sharpen a mixed two-modality feature, then use the enhanced mixed feature to sharpen each modality again, and this two-way exchange reduces interference between RGB and IR. Adding a bi-directional decoupled focus to the backbone enlarges the receptive field in horizontal and vertical directions. On M3FD and LLVIP, the authors report that this combination outperforms state-of-the-art single-modality detectors and fusion-and-detection methods by a clear margin, with DEYOLO-l reaching 91.2 mAP50 on M3FD and DEYOLO-n 96.8 mAP50 on LLVIP.

Load-bearing premise

The central assumption is that the validation split used for all headline comparisons is an honest measure of detection quality, even though that same split was used to choose module hyperparameters and no baseline was scored on the held-out test portion.

Editorial extensions

If this is right

  • On M3FD, adding DECA, DEPA, and the bi-directional focus lifts YOLOv8n from 80.8 to 86.6 mAP50; on LLVIP, DEYOLO-n reaches 96.8 mAP50, above all listed fusion-and-detection baselines.
  • A detector can exploit infrared information without a separate image-fusion network, so training and inference become a single end-to-end detection model.
  • DECA and DEPA are described as plug-and-play, meaning other detection backbones could adopt the same dual-enhancement fusion pattern.
  • The reported KAIST results suggest the approach also applies to RGB-thermal detection, though not to state-of-the-art level there.
  • The dual-enhancement logic implies that mutual interference between modalities can be handled in feature space, not only at the image level.

Reading between the lines

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

  • If the validation gains survive on a true held-out test set, the practical takeaway is that feature-level, detection-oriented fusion is a cheaper and more direct route than image-fusion-then-detect pipelines for low-light RGB-IR systems.
  • The same dual-enhancement pattern could be tested on other paired inputs, such as RGB-depth or multi-spectral images, where one modality is reliable and the other is complementary.
  • A reader should treat the reported margins as validation-split estimates: the paper selects DEPA kernel size and DECA depth on the same split, so independent replication on a different split or dataset is the natural next check.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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 DEYOLO, a YOLOv8-based RGB-infrared object detector. It introduces a dual semantic enhancing channel weight assignment module (DECA), a dual spatial enhancing pixel weight assignment module (DEPA), and a bi-directional decoupled focus module in the backbone. The method is evaluated on M3FD and LLVIP under a self-defined 3:1:1 split, with additional experiments on KAIST, and the paper claims that DEYOLO outperforms state-of-the-art object detectors and fusion-and-detection methods by a clear margin.

Significance. If the empirical claims were fully supported, the paper would make a useful contribution: detection-centered cross-modality fusion is a practical direction, the proposed modules are simple and plausibly plug-and-play, and the ablation in Table 1 shows consistent gains from each component. The code release is also a concrete reproducibility asset. However, the current evidence does not support the stated 'clear margin' over single-modality state-of-the-art detectors, and the evaluation protocol needs substantial strengthening before the headline claim can be accepted.

major comments (4)
  1. [Abstract; Section 4.4, Table 4] The headline claim that DEYOLO 'outperforms SOTA object detection algorithms by a clear margin' is contradicted by the paper's own Table 4. DEYOLO-n obtains 86.6 mAP50 / 58.9 mAP50-95, while single-modality YOLOv7 (visible) obtains 90.4 / 61.3 and YOLOv8l (visible) obtains 88.3 / 61.8. Only DEYOLO-l (91.2 / 66.3) tops the table, and for that comparison the table does not report parameter-matched baselines, per-model parameter counts, or a YOLOv8l cross-modality row. Please either add the missing baselines and complexity measures or revise the 'clear margin' and 'outperforms all single-modality' claims.
  2. [Section 4.2-4.4, Tables 2-5] All headline numbers in Tables 4-5 are mAP on a self-defined 3:1:1 validation split, and the same split was used to select DEPA's kernel size (Table 2), DECA's layer type and count (Table 3), and module placement. The test split is mentioned only in one sentence reporting DEYOLO's test mAP, with no baseline evaluated on it. This makes the claimed margins selection-dependent; please report test-split results for all compared methods or use the official train/test splits.
  3. [Section 4.3, Table 1] The ablation baseline is 'the baseline network trained merely by visible images' (80.8 / 54.3), while the DEYOLO rows use both modalities. Since the improvements co-occur with the addition of the infrared stream, Table 1 does not isolate the contribution of DECA, DEPA, and the bi-direction focus from the benefit of simply having a second modality. Please add a cross-modality baseline without the proposed modules, such as YOLOv8n with a four-channel input or with two-branch late fusion, and report its mAP in Table 1.
  4. [Section 4.4, Table 5] The margin over fusion-and-detection methods is not 'clear' for the small model: on LLVIP, DEYOLO-n leads the next best method (SeAFusion) by only 0.6 mAP50, and on M3FD the stated minimum gains of 5.4 / 3.1 points are computed only against the fusion-and-detection baselines, not against the single-modality detectors in Table 4. In addition, the fusion baselines are not retrained on the same 3:1:1 split or with the same protocol, so the comparison may conflate detector training choices with fusion quality. Please clarify the training protocol for all baselines and temper the margin claims accordingly.
minor comments (6)
  1. [Section 1, Fig. 1 caption] The caption contains the typo 'U2Fsuion' for 'U2Fusion'; please fix it.
  2. [Section 3.2, Eq. (7)] The convolution operations conv1 and conv2 in Eq. (7) are not defined in the method text; the default kernel size is only inferable from Table 2. Please specify these operations explicitly.
  3. [Section 3.2, Eq. (6)] The phrase 'an element-wise multiplication is applied on the result of each other' is unclear; please state explicitly that the two convolved features are multiplied element-wise.
  4. [Section 4.2] The text says the authors use Log Average Miss Rate (LAMR) as an evaluation metric, but no LAMR values are reported anywhere in the paper; please add them or remove the sentence.
  5. [Table 4] The YOLOv8l row is missing a cross-modality entry, and no parameter counts or FLOPs are provided for any model; please complete the table to make the comparisons interpretable.
  6. [Figure 3] Figure 3 labels parts of DECA as 'Linear' while the text describes convolutions (Eqs. (2)-(4)); please align the figure with the text.

Circularity Check

1 steps flagged · score 4.0 of 10

No algebraic circularity; the main circularity concern is that DEYOLO's headline mAPs are measured on the same validation split used to select DECA/DEPA hyperparameters, with no baseline run on the held-out test split.

  1. fitted input called prediction [Section 4.2 (evaluation protocol), Section 4.3 Tables 2-3, Section 4.4 Tables 4-5]
    "And, we evaluate our method on the validation set and use the mean average precision (mAP) with the IoU threshold of 0.5 and Log Average Miss Rate (LAMR) as the evaluation metric. ... We have re-split the datasets into training, validation, and test sets in a 3:1:1 ratio. After dividing the test set as described above, the mAP50 on the test/validation sets of the two datasets are 85.7%/86.6% and 96.4%/96.8%, respectively."

    Tables 2 and 3 select DEPA's kernel size and DECA's layer count by comparing validation-set mAP on M3FD. Tables 4 and 5 then present DEYOLO's headline numbers (e.g., 86.6/58.9 on M3FD) on that same validation split, and the abstract/Section 4.4 convert these into the claim of a 'clear margin' over SOTA. Because the validation split was used to choose the modules' hyperparameters, the reported margin is a fitted summary of the selection data rather than an unbiased estimate on a held-out set. No baseline is evaluated on the test split, so the central comparison is statistically dependent on the same data used for model selection, even though it is not an algebraic identity.

full rationale

DEYOLO is an empirical detector paper: the DECA/DEPA equations define a fusion architecture, and the claimed contribution is measured detection accuracy, not a quantity derived from the architecture by construction. There is no uniqueness theorem, no ansatz smuggled in via citation, and no known result merely renamed. The paper does contain self-citations (SeAFusion and PIAFusion list Yuan, a co-author), but those are used as comparison baselines and are not load-bearing justifications for DEYOLO's design, so they do not by themselves raise the circularity score. The principal circularity concern is evaluation-protocol-level: the DEPA kernel size and DECA layer count are tuned on the validation split (Tables 2-3), and the same split's mAP is then reported as the SOTA-beating result in Tables 4-5 and the abstract. The test split numbers (85.7/96.4) are reported only for DEYOLO, with no baseline comparison, so the 'clear margin' claim rests on fitted validation numbers. Separately, Section 4.4's assertion that 'DEYOLO outperforms all single-modality object detection models' is contradicted by Table 4 for DEYOLO-n, since YOLOv7-visible (90.4/61.3) and YOLOv8l-visible (88.3/61.8) exceed DEYOLO-n (86.6/58.9); this is a correctness/evidence problem rather than circularity. Because the derivation itself is not circular but the headline comparison is statistically tied to the hyperparameter search on the same split, a score of 4 is appropriate rather than 0 or 6+.

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

This is an empirical architecture paper. There is no mathematical derivation of the claimed improvements; the only hand-tuned quantities are architectural hyperparameters selected on the validation set. The paper assumes the standard benchmarks, the YOLOv8 base training protocol, and the effectiveness of attention-style weighting. No new physical or conceptual entities are introduced.

free parameters (3)
  • DEPA mixed-feature convolution kernel size = 3x3
    Table 2 compares 3x3, 5x5, and 7x7 and selects 3x3 based on validation mAP50 (85.3), so the final model's design is tuned to the evaluation metric.
  • DECA cross-modality weight extraction layers and type = 3 depthwise-conv layers
    Table 3 compares 2 and 3 normal conv layers and 2 and 3 depthwise conv layers; 3 depthwise layers achieve the best validation mAP50 (85.2), and one layer fails to converge.
  • Number and placement of DECA/DEPA blocks across scales = unspecified
    The text says DECAs and DEPAs are added between backbone and necks at multi-scale features, but the exact scales and counts are not stated; these choices affect the results.
assumptions (3)
  • domain assumption M3FD, LLVIP, and cleaned KAIST annotations are correct, pixel-aligned, and compatible with COCO-style mAP evaluation
    The paper relies on these public datasets for all training and evaluation; no manual verification or error analysis is provided.
  • domain assumption YOLOv8 baseline and the compared fusion models are used in a way that isolates the proposed modules' contribution
    The ablation adds modules to YOLOv8n and measures gains; this assumes YOLOv8's training setup is a neutral base and the baselines are not disadvantaged by the comparison protocol.
  • domain assumption Attention-based reweighting (SE-style channel weights, softmax spatial masks) improves detection when applied cross-modality
    The core mechanism of DECA and DEPA is motivated by SE blocks and spatial attention; the paper provides no proof, only empirical mAP gains on validation.

how reviews work

0 comments
Cite this review

Pith. "Pith review of DEYOLO: Dual-Feature-Enhancement YOLO for Cross-Modality Object Detection." pith.science (2026). https://pith.science/paper/W3NH7FYH

@misc{pith2026241204931,
  author       = {Pith},
  title        = {Pith review of: DEYOLO: Dual-Feature-Enhancement YOLO for Cross-Modality Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/W3NH7FYH}},
  note         = {Machine review of arXiv:2412.04931}
}
read the original abstract

Object detection in poor-illumination environments is a challenging task as objects are usually not clearly visible in RGB images. As infrared images provide additional clear edge information that complements RGB images, fusing RGB and infrared images has potential to enhance the detection ability in poor-illumination environments. However, existing works involving both visible and infrared images only focus on image fusion, instead of object detection. Moreover, they directly fuse the two kinds of image modalities, which ignores the mutual interference between them. To fuse the two modalities to maximize the advantages of cross-modality, we design a dual-enhancement-based cross-modality object detection network DEYOLO, in which semantic-spatial cross modality and novel bi-directional decoupled focus modules are designed to achieve the detection-centered mutual enhancement of RGB-infrared (RGB-IR). Specifically, a dual semantic enhancing channel weight assignment module (DECA) and a dual spatial enhancing pixel weight assignment module (DEPA) are firstly proposed to aggregate cross-modality information in the feature space to improve the feature representation ability, such that feature fusion can aim at the object detection task. Meanwhile, a dual-enhancement mechanism, including enhancements for two-modality fusion and single modality, is designed in both DECAand DEPAto reduce interference between the two kinds of image modalities. Then, a novel bi-directional decoupled focus is developed to enlarge the receptive field of the backbone network in different directions, which improves the representation quality of DEYOLO. Extensive experiments on M3FD and LLVIP show that our approach outperforms SOTA object detection algorithms by a clear margin. Our code is available at https://github.com/chips96/DEYOLO.

Figures

Figures reproduced from arXiv: 2412.04931 by the authors.

Figure 1
Figure 1. Detection results of different methods. 2 Related Work In this section, we review the commonly used single-modality object detection algorithms first. Then, some recent visible and infrared image fusion methods are introduced. 2.1 Single-Modality Object Detection Recently, deep neural networks have been proposed to improve accuracy in ob￾ject detection tasks, including CNN and its variants, e.g. Sparse R-CNN [23], C… view at source ↗
Figure 2
Figure 2. The framework of the proposed DEYOLO. We incorporate dual-context col￾laborative enhancement modules (DECA and DEPA) within the feature extraction streams dedicated to each detection head in order to refine the single-modality features and fuse multi-modality representations. Concurrently, the Bi-direction Decoupled Fo￾cus is inserted in the early layers of the YOLOv8 backbone to expand the network’s receptive field… view at source ↗
Figure 3
Figure 3. The concrete structure of DECA and DEPA. These modules utilize both single￾modality and cross-modality information through a dual enhancement mechanism. DECA enhances the cross-modality fusion results by leveraging dependencies between channels within each modality and outcomes are then used to reinforce the original single-modal features, highlighting more discriminative channels. Similarly, DEPA is able to learn d… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Bi-direction decoupled focus. 3.3 Bi-direction Decoupled Focus In this subsection, we tend to improve the performance of object detection from the perspective of the single modality. In order to enhance the capability of extracting targets, the bi-direction decoupled f…
Figure 6
Figure 6. Figure 6 [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 26 canonical work pages

  1. [1]

    https://www.flir.in/oem/ adas/adas-dataset-form (2018)

    FLIR: Flir thermal dataset for algorithm training. https://www.flir.in/oem/ adas/adas-dataset-form (2018)

    work page 2018

  2. [2]

    arXiv preprint arXiv:2004.10934 (2020)

    Bochkovskiy, A., Wang, C.Y., Liao, H.Y.M.: Yolov4: Optimal speed and accuracy of object detection. arXiv preprint arXiv:2004.10934 (2020)

    arXiv 2020

  3. [3]

    In: European conference on computer vision

    Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., Zagoruyko, S.: End- to-end object detection with transformers. In: European conference on computer vision. pp. 213–229. Springer (2020) Dual-Feature-Enhancement YOLO 15

    work page 2020

  4. [4]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Chollet, F.: Xception: Deep learning with depthwise separable convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1251–1258 (2017)

    2017

  5. [5]

    IEEE Transactions on Geoscience and Remote Sensing 59(11), 9813–9824 (2021)

    Dai, Y., Wu, Y., Zhou, F., Barnard, K.: Attentional local contrast networks for infrared small target detection. IEEE Transactions on Geoscience and Remote Sensing 59(11), 9813–9824 (2021)

    work page 2021

  6. [6]

    Infor- mation Fusion50, 148–157 (2019)

    Guan, D., Cao, Y., Yang, J., Cao, Y., Yang, M.Y.: Fusion of multispectral data through illumination-aware deep neural networks for pedestrian detection. Infor- mation Fusion50, 148–157 (2019)

    work page 2019

  7. [7]

    IEEE Geoscience and Remote Sensing Letters 19, 1–5 (2022)

    Hou, Q., Zhang, L., Tan, F., Xi, Y., Zheng, H., Li, N.: Istdu-net: Infrared small- target detection u-net. IEEE Geoscience and Remote Sensing Letters 19, 1–5 (2022). https://doi.org/10.1109/LGRS.2022.3141584

    work page arXiv 2022

  8. [8]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 7132–7141 (2018)

    2018

Show all 36 references
  1. [9]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Hwang, S., Park, J., Kim, N., Choi, Y., So Kweon, I.: Multispectral pedestrian detection: Benchmark dataset and baseline. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 1037–1045 (2015)

    2015

  2. [10]

    In: Proceedings of the IEEE/CVF international con- ference on computer vision

    Jia, X., Zhu, C., Li, M., Tang, W., Zhou, W.: LLVIP: A visible-infrared paired dataset for low-light vision. In: Proceedings of the IEEE/CVF international con- ference on computer vision. pp. 3496–3504 (2021)

    2021

  3. [11]

    https://doi.org/10.5281/ zenodo.3908559, https://github.com/ultralytics/yolov5

    Jocher, G.: YOLOv5 by Ultralytics (May 2020). https://doi.org/10.5281/ zenodo.3908559, https://github.com/ultralytics/yolov5

    2020

  4. [12]

    https://github.com/ultralytics/ultralytics (2024)

    Jocher, G.: ultralytics/yolov8: v8.1.0 - yolov8 oriented bounding boxes (obb). https://github.com/ultralytics/ultralytics (2024)

    2024

  5. [13]

    In: European conference on computer vision

    Kieu, M., Bagdanov, A.D., Bertini, M., Del Bimbo, A.: Task-conditioned domain adaptation for pedestrian detection in thermal imagery. In: European conference on computer vision. pp. 546–562. Springer (2020)

    2020

  6. [14]

    Pattern Recognition85, 161–171 (2019)

    Li, C., Song, D., Tong, R., Tang, M.: Illumination-aware faster R-CNN for robust multispectral pedestrian detection. Pattern Recognition85, 161–171 (2019)

    2019

  7. [15]

    arXiv preprint arXiv:1611.02644 (2016)

    Liu, J., Zhang, S., Wang, S., Metaxas, D.N.: Multispectral deep neural networks for pedestrian detection. arXiv preprint arXiv:1611.02644 (2016)

    2016 arXiv

  8. [16]

    In: Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern Recognition

    Liu, J., Fan, X., Huang, Z., Wu, G., Liu, R., Zhong, W., Luo, Z.: Target-aware dual adversarial learning and a multi-scenario multi-modality benchmark to fuse infrared and visible for object detection. In: Proceedings of the IEEE/CVF Con- ference on Computer Vision and Pattern...

    2022

  9. [17]

    IEEE Transactions on Circuits and Systems for Video Technology32(1), 105–119 (2021)

    Liu, J., Fan, X., Jiang, J., Liu, R., Luo, Z.: Learning a deep multi-scale feature ensemble and an edge-attention guidance for image fusion. IEEE Transactions on Circuits and Systems for Video Technology32(1), 105–119 (2021)

    2021

  10. [18]

    Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B.: Swin transformer:Hierarchical visiontransformerusingshiftedwindows.In:Proceedings of the IEEE/CVF international conference on computer vision. pp. 10012–10022 (2021)

    2021

  11. [19]

    IEEE/CAA Journal of Automatica Sinica9(7), 1200–1217 (2022)

    Ma, J., Tang, L., Fan, F., Huang, J., Mei, X., Ma, Y.: SwinFusion: Cross-domain long-range learning for general image fusion via swin transformer. IEEE/CAA Journal of Automatica Sinica9(7), 1200–1217 (2022)

    2022

  12. [20]

    Pattern Recognition80, 143–155 (2018)

    Park, K., Kim, S., Sohn, K.: Unified multi-spectral pedestrian detection based on probabilistic fusion networks. Pattern Recognition80, 143–155 (2018)

    2018

  13. [21]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Redmon, J., Divvala, S., Girshick, R., Farhadi, A.: You only look once: Unified, real-time object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 779–788 (2016) 16 Yishuo et al

    2016

  14. [22]

    In: Proceedings of the IEEE conference on computer vision and pattern recognition

    Redmon, J., Farhadi, A.: YOLO9000: better, faster, stronger. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 7263–7271 (2017)

    2017

  15. [23]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

    Sun, P., Zhang, R., Jiang, Y., Kong, T., Xu, C., Zhan, W., Tomizuka, M., Li, L., Yuan, Z., Wang, C., et al.: Sparse r-cnn: End-to-end object detection with learnable proposals. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 14454–144...

    2021

  16. [24]

    In: Proceedings of the 30th ACM International Conference on Multimedia

    Sun, Y., Cao, B., Zhu, P., Hu, Q.: Detfusion: A detection-driven infrared and visible image fusion network. In: Proceedings of the 30th ACM International Conference on Multimedia. pp. 4003–4011 (2022)

    2022

  17. [25]

    Information Fusion 82, 28–42 (2022)

    Tang, L., Yuan, J., Ma, J.: Image fusion in the loop of high-level vision tasks: A semantic-aware real-time infrared and visible image fusion network. Information Fusion 82, 28–42 (2022)

    2022

  18. [26]

    Information Fusion 83, 79–92 (2022)

    Tang, L., Yuan, J., Zhang, H., Jiang, X., Ma, J.: PIAFusion: A progressive infrared and visible image fusion network based on illumination aware. Information Fusion 83, 79–92 (2022)

    2022

  19. [27]

    Data in brief15, 249–251 (2017)

    Toet, A.: The TNO multiband image data collection. Data in brief15, 249–251 (2017)

    2017

  20. [28]

    Advances in neural information pro- cessing systems 30 (2017)

    Vaswani,A.,Shazeer,N.,Parmar,N.,Uszkoreit,J.,Jones,L.,Gomez,A.N.,Kaiser, Ł., Polosukhin, I.: Attention is all you need. Advances in neural information pro- cessing systems 30 (2017)

    2017

  21. [29]

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

    Wang, C.Y., Bochkovskiy, A., Liao, H.Y.M.: YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 7464– 7475 (2023)

    2023

  22. [30]

    Information Fusion98, 101828 (2023)

    Wang, D., Liu, J., Liu, R., Fan, X.: An interactively reinforced paradigm for joint infrared-visible image fusion and saliency object detection. Information Fusion98, 101828 (2023)

    2023

  23. [31]

    IEEE Transactions on Pattern Analysis and Machine Intelligence 44(1), 502–518 (2020)

    Xu, H., Ma, J., Jiang, J., Guo, X., Ling, H.: U2fusion: A unified unsupervised image fusion network. IEEE Transactions on Pattern Analysis and Machine Intelligence 44(1), 502–518 (2020)

    2020

  24. [32]

    In: proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence (2020)

    Xu, H., Ma, J., Le, Z., Jiang, J., Guo, X.: FusionDN: A Unified Densely Connected Network for Image Fusion. In: proceedings of the Thirty-Fourth AAAI Conference on Artificial Intelligence (2020)

    2020

  25. [33]

    In- formation Fusion50, 20–29 (2019)

    Zhang, L., Liu, Z., Zhang, S., Yang, X., Qiao, H., Huang, K., Hussain, A.: Cross- modality interactive attention network for multispectral pedestrian detection. In- formation Fusion50, 20–29 (2019)

    2019

  26. [34]

    IEEE Transactions on Geoscience and Remote Sensing59(5), 4481–4492 (2020)

    Zhao, B., Wang, C., Fu, Q., Han, Z.: A novel pattern for infrared small target detection with generative adversarial network. IEEE Transactions on Geoscience and Remote Sensing59(5), 4481–4492 (2020)

    2020

  27. [35]

    In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR)

    Zhao, Z., Bai, H., Zhang, J., Zhang, Y., Xu, S., Lin, Z., Timofte, R., Van Gool, L.: CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi- Modality Image Fusion. In: Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR). ...

    2023

  28. [36]

    arXiv preprint arXiv:2103.07461 (2021)

    Zhou, X., Koltun, V., Krähenbühl, P.: Probabilistic two-stage detection. arXiv preprint arXiv:2103.07461 (2021)

    2021 arXiv

Pith tools

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