Pith. sign in

REVIEW 5 major objections 6 minor 40 references

Rethinking Features-Fused-Pyramid-Neck for Object Detection

T0 review · 5 major / 6 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read FPN-style feature fusion is misaligned; a scalar reweighting fixes it.

desk verdict A broad empirical study of FPN alternatives that overreaches on mechanism: the SNI scalar is likely a reparameterization of the conv-BN, so the 'feature misalignment' story is unsupported, but the module-level gains may still be useful. read the letter →

arxiv 2505.12820 v1 pith:3UISV555 submitted 2025-05-19 cs.CV

Hulin Li This is my paper · ORCID
classification cs.CV
keywords objectdetectionfeaturepyramidmisalignmentsoftnearestneighborinterpolationdownsamplinglightweightconvolutionreal-timeYOLO
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that the standard practice of fusing feature maps from different pyramid levels point-to-point in detection necks is fundamentally flawed: low-level features emphasize small and local objects while high-level features emphasize large and global ones, so merging them directly injects noise. To validate this, it builds an unfused independent hierarchy pyramid (IHP) neck that beats fused baselines on coupled-head detectors, then introduces soft nearest neighbor interpolation (SNI), a single scalar multiplication on upsampled high-level features that shrinks their contribution as the resolution gap grows. Combined with a new downsampling module (ESD) and a lightweight convolution (GSConvE), these form the secondary features alignment (SA) solution tested on six YOLO variants. The claim is that this nearly cost-free modification improves accuracy across all tested models and achieves state-of-the-art results on Pascal VOC and MS COCO, suggesting that elaborate fusion architectures may be doing more harm than good.

What carries the argument

The load-bearing mechanism is a pair of scale-dependent reweighting operations at the two places where feature resolution changes. SNI (soft nearest neighbor interpolation) replaces plain nearest-neighbor upsampling with $Y = \alpha \cdot f(X)$, $\alpha = \frac{\text{Resolution}_X}{\text{Resolution}_Y}$, a scalar that weakens high-level features as the zoom factor grows while keeping nearest-neighbor speed. ESD (features adaptive selection in extended spatial windows) replaces stride-2 downsampling with three parallel branches — a $3\times3$ convolution, a $4\times4$ max-pooling, and a $4\times4$ average-pooling — merged by addition (ESD-I) or learnable linear fusion (ESD-II) to preserve spatial detail. GSConvE simplifies GSConv into variants with large-kernel depthwise branches ($9\times9$, $13\times13$, $17\times17$) for cheap large receptive fields, and IHP (independent hierarchy pyramid) is the diagnostic neck with no fusion, whose behavior isolates the effect of misalignment.

What would settle it

Train a fixed detector (for example YOLOv5-n) with and without SNI while holding parameter counts equal, and sweep $\alpha$ from 0 to 1: if accuracy does not peak near $\alpha = \frac{\text{Resolution}_X}{\text{Resolution}_Y}$, or if the gain disappears at $\alpha = 1$, the misalignment explanation fails. Applying SNI to a neck that fuses spatially aligned features (for instance after a learned offset or deconvolution) would further test whether the improvement is specific to misaligned fusion.

Watch

Extended reading notes

Core claim

The central discovery is that feature misalignment in FPN-like necks is systematic, and that a fixed scalar reweighting can mitigate it. For every upsampled feature map $X$ fused into a higher-resolution map $Y$, the paper defines SNI as $Y = \alpha \cdot f(X)$ with $\alpha = \frac{\text{Resolution}_X}{\text{Resolution}_Y}$, so high-level features are downweighted in proportion to how much they are enlarged. Ablations on VOC 07+12 show SNI raising accuracy for every one of the six YOLO baselines, with gains up to +3.3 AP for YOLOv7-tiny and +2.0 AP for YOLOv6-n at no added parameters or FLOPs, and the full SA solution reaches the best accuracy among real-time detectors on both VOC and COCO. The paper also shows that dropping fusion entirely (IHP) helps coupled-head detectors but hurts decoupled-head ones, which it explains by decoupled heads already separating classification and localization features.

Load-bearing premise

The accuracy gains are attributed to the proposed neck components even though some compared models differ in parameter count and FLOPs, and the SNI downweighting factor $\alpha$ is fixed by formula without a sensitivity study.

Editorial extensions

If this is right

  • Detectors already deployed with FPN-style necks can gain accuracy by simply swapping the nearest-neighbor upsampling for SNI and keeping everything else fixed.
  • The IHP result implies that feature fusion is not an obligatory component for coupled-head detectors, so future architecture search can treat fusion as optional.
  • The ESD result points to downsampling as a recoverable source of spatial information loss, worth revisiting in backbone design beyond the neck.
  • Consistent gains across six different baselines suggest the misalignment problem is architectural, not specific to one training recipe or dataset.

Reading between the lines

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

  • If SNI acts primarily as a size-dependent gradient regularizer rather than an alignment device, it should also change results when fusing already-aligned feature maps; testing that case would separate the two explanations (the paper does not run this control).
  • A sensitivity sweep over $\alpha$ near $\frac{\text{Resolution}_X}{\text{Resolution}_Y}$ could show whether the exact ratio is optimal or merely a good default; a learned per-layer scalar that matches or beats it would suggest the mechanism is adaptive scaling, not the specific formula.
  • The same misalignment argument should apply to semantic and panoptic segmentation, where boundary regions between object scales are denser, so the SA modules may transfer to those tasks, though the paper only evaluates detection benchmarks.
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

5 major / 6 minor

Summary. The paper argues that point-to-point fusion of feature maps from different pyramid levels in FPN-style necks causes feature misalignment, and that this hurts real-time object detectors. To address it, the authors propose an independent hierarchy pyramid (IHP) neck that removes fusion, a soft nearest neighbor interpolation (SNI) that down-weights upsampled high-level features by a fixed scalar alpha = ResolutionX / ResolutionY, a features adaptive selection downsampling method (ESD), and a lightweight convolution module (GSConvE). These components are combined into a 'secondary features alignment' (SA) solution and evaluated on Pascal VOC and MS COCO across six YOLO baselines, with claims of state-of-the-art results. The paper also reports a comparison of confidence thresholds to motivate a more 'realistic' evaluation protocol.

Significance. The empirical breadth is a genuine strength: the IHP, SNI, ESD, and GSConvE are tested across six YOLO baselines, the SNI gains are consistent, and the architecture table plus the promise of released code would aid reproducibility. If the claimed effects are real, a zero-cost scalar multiplier on upsampled features plus simple downsampling and convolution changes would be a practically useful recipe for real-time detectors. However, the central mechanistic claim about feature misalignment is not established by the experiments as designed, and the headline comparisons are not parameter- or FLOP-matched. The paper would be significantly strengthened by control experiments that separate capacity effects from the proposed mechanisms.

major comments (5)
  1. [Sec. 3.2, Eq. (1)] The 'soft nearest neighbor interpolation' is a single scalar alpha = ResolutionX / ResolutionY multiplied onto the entire upsampled feature map. Because every spatial location and every channel receives the same factor, it cannot change which features are spatially aligned. In the PANet-style necks used here, the upsampled branch is concatenated or added to a lateral branch and then passed through a convolution followed by BatchNorm; a constant scaling of one input branch is exactly equivalent to rescaling the corresponding weights of that convolution, with BatchNorm absorbing the scale. The representable function class is therefore unchanged, so the consistent gains in Table 1 Part II do not by themselves confirm the existence of feature misalignment. The authors should add a learned-scalar baseline (e.g., a per-channel or global learned alpha), a sensitivity study over alpha, and a reparameterization check to show that the gain is not simply an initialization or optimization effect.
  2. [Sec. 4.3, Tables 4-6] The headline state-of-the-art comparisons are not matched in parameters or FLOPs. For example, SYOLO has 57.2M parameters and 142.5 GFLOPs, while YOLOv8 has 43.7M and 165.2 GFLOPs; SYolo-s has 6.6M parameters while YOLOv8-n has 3.2M. Differences of this size mean the accuracy gaps could be due to extra capacity rather than the proposed SNI, ESD, and GSConvE components. The authors should report matched-capacity baselines, or ablations in which the SA components are swapped into the exact same backbone-neck-head with the same parameter and FLOP budget.
  3. [Table 1, Yolov3-t-esd-I row] The reported gain for Yolov3-t-esd-I is 11.2 AP and 11.2 AP50 (from 58.0 to 69.2 AP50 and 27.3 to 38.5 AP), while the parameter count grows from 8.71M to 11.8M and FLOPs from 13.0G to 22.4G. This single outlying result is far larger than all other ESD gains and is not discussed in the text. It suggests either a configuration difference, an extra-capacity effect, or a possible reporting error. The aggregate claim that ESD improves accuracy is not supported until this row is explained or corrected.
  4. [Sec. 4.2, Tables 1-2] All ablations are single runs, and many differences are 0.1-0.3 AP, which is within the typical run-to-run noise of object detection experiments. At least three seeds with means and standard deviations should be reported for the key comparisons, especially the SNI and ESD claims, so that the reader can judge whether the improvements are statistically meaningful.
  5. [Sec. 4.3 and Table 7] The final SYOLO architecture is never ablated as a whole from a standard baseline. Because SYOLO changes the backbone (E-ELAN/C2f), the neck (SNI, E-ELAN, GSConvE), and the downsampling (ESD) simultaneously, the final COCO result cannot be attributed to the proposed components. A cumulative ablation (baseline -> +SNI -> +ESD -> +GSConvE -> full SYOLO) with matched or explicitly reported complexity is needed to support the claim that the SA solution causes the state-of-the-art result.
minor comments (6)
  1. [Abstract/header] The phrase 'This paper has been accepted byECCV2024 and published on Springer Nature' is a provenance note rather than scientific content; if this is the published version, cite the ECCV proceedings and clarify the relationship between the arXiv preprint and the published version.
  2. [Table 6] The citation [37] is given for PPYoloE, but reference [37] is titled '3D-Man' and appears unrelated; please correct the citation.
  3. [Sec. 3.3] The terms 'norm models' and 'lightweight models' are used to choose between ESD-I and ESD-II and between GSConvE-I and GSConvE-II, but they are never defined; give a concrete criterion such as a parameter count or FLOPs threshold.
  4. [Sec. 4.1] The description of training hyperparameters as 'the default set of hyperparameters' is insufficient for reproducibility; please report epochs, batch size, optimizer, learning rate schedule, and augmentation settings for the ablations.
  5. [Table 3 and Sec. 4.2] The comparison between confidence thresholds 0.001 and 0.25 is presented as a 'realistic assessment,' but AP is conventionally computed over a precision-recall curve that includes low-confidence detections; please state the exact evaluation protocol, including whether NMS settings were identical, so readers can interpret the reported AP drop.
  6. [Table 4 caption] The caption contains a typo: 'Comparasion' should be 'Comparison'.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the core components are empirically evaluated against external baselines, and the SNI scaling factor is defined from input resolutions rather than fitted to the reported detection results.

full rationale

The paper's central claims are empirical rather than definitional. The SNI operation is defined in Eq. (1) as Y = α·f(X) with α = ResolutionX / ResolutionY, where the scaling factor is computed directly from the two feature-map resolutions and is not fitted to any detection accuracy, so the reported gains are not a renamed fit. The load-bearing comparisons in Tables 1, 4, and 5 are against standard external baselines (YOLOv3 through YOLOv8) with fixed hyperparameters and no pre-trained weights, which gives the accuracy claims independent empirical content. The GSConvE section builds on the author's own GSConv paper [18], but that citation is used only to describe the antecedent module being modified; the GSConvE-I and GSConvE-II ablations are measured against the published GSConv baselines, so the self-citation is descriptive and not load-bearing. No uniqueness theorem is imported, no input quantity is renamed as a prediction, and no ansatz is smuggled in via a prior self-citation. The skeptic's argument that a spatially uniform scalar on the upsampled branch can be absorbed into the subsequent convolution and batch normalization is a plausible alternative explanation for the empirical gains, and it may indicate that the mechanistic claim about feature misalignment is under-supported, but it is not circular: the experiments do not derive the misalignment conclusion from the definition of SNI. Concerns about parameter-count differences and the absence of a sensitivity study for α are correctness or completeness issues, not circularity. The derivation chain is therefore self-contained with respect to the identified circularity patterns.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical entities or formal axioms. The ledger lists the hand-chosen architecture hyperparameters and the empirical assumptions that the reported gains depend on.

free parameters (3)
  • ESD extended window size for max and average pooling = 4x4
    Chosen in Fig. 4 and used in Table 7; no sensitivity analysis is provided, and it directly affects the spatial information retained during downsampling.
  • GSConvE-II depthwise kernel sizes = 9x9, 13x13, 17x17
    Selected in Section 3.4 for the auxiliary branches; the paper does not ablate other kernel sizes or justify this particular range.
  • ESD fusion mode assignment = I for lightweight models, II for norm models
    Assigned by model scale rather than by a measured rule; Table 1 shows different gains for ESD-I and ESD-II across baselines.
assumptions (5)
  • domain assumption Representation bias across pyramid levels causes point-to-point fusion to inject noise into detection features.
    Used to motivate IHP and SNI in Sections 1 and 3.1; supported only by qualitative heatmaps in Fig. 1, not by a quantitative proof.
  • domain assumption A scalar downweighting of upsampled high-level features is sufficient to fix misalignment.
    Eq. (1) defines alpha with no comparison against learned fusion weights or other constants.
  • domain assumption Ablation results on Pascal VOC 07+12 transfer to MS COCO.
    Component ablations in Tables 1 and 2 use VOC, while headline claims use COCO; transfer of hyperparameters is assumed.
  • domain assumption Larger depthwise kernels in GSConvE capture global context without requiring more layers.
    Motivates the 9x9 to 17x17 kernels in Section 3.4; no theoretical or ablative support is given for this particular range.
  • standard math Standard CNN training, backpropagation, and default hyperparameters from baseline repositories are reliable.
    The paper relies on standard deep learning machinery and does not re-derive it; this is not a point of contention.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Rethinking Features-Fused-Pyramid-Neck for Object Detection." pith.science (2026). https://pith.science/paper/3UISV555

@misc{pith2026250512820,
  author       = {Pith},
  title        = {Pith review of: Rethinking Features-Fused-Pyramid-Neck for Object Detection},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3UISV555}},
  note         = {Machine review of arXiv:2505.12820}
}
read the original abstract

Multi-head detectors typically employ a features-fused-pyramid-neck for multi-scale detection and are widely adopted in the industry. However, this approach faces feature misalignment when representations from different hierarchical levels of the feature pyramid are forcibly fused point-to-point. To address this issue, we designed an independent hierarchy pyramid (IHP) architecture to evaluate the effectiveness of the features-unfused-pyramid-neck for multi-head detectors. Subsequently, we introduced soft nearest neighbor interpolation (SNI) with a weight downscaling factor to mitigate the impact of feature fusion at different hierarchies while preserving key textures. Furthermore, we present a features adaptive selection method for down sampling in extended spatial windows (ESD) to retain spatial features and enhance lightweight convolutional techniques (GSConvE). These advancements culminate in our secondary features alignment solution (SA) for real-time detection, achieving state-of-the-art results on Pascal VOC and MS COCO. Code will be released at https://github.com/AlanLi1997/rethinking-fpn. This paper has been accepted by ECCV2024 and published on Springer Nature.

Figures

Figures reproduced from arXiv: 2505.12820 by the authors.

Figure 1
Figure 1. The IHP and seven other typical neck architectures. The three heat maps are come from the third (P3), fourth (P4), and fifth (P5) hierarchies of the FPN-neck model, respectively. The area of interest for each hierarchies’ features is highlighted in red; the darker the color, the higher the feature weight. It is evident that lower-level hierarchies prioritize local features, while higher-level hierarchies focus more … view at source ↗
Figure 2
Figure 2. Illustrations of the features misalignment in fusion and the SNI. These numbers, 22-88, are just markers of different local-features not real feature values. P3 P4 P5 soft weight, retain texture identical input Model with SNI Model with traditional method output (no SNI) output How to fuse? P3 P4 P5 soft weight, retain texture identical input Model with SNI Model with traditional method output (no SNI) output How to… view at source ↗
Figure 3
Figure 3. A comparison of the SNI and traditional method. This result could be repro￾duced by using SNI source code (model:Yolov5n-panet). The same size scaling before and after up-sampling is for intuitive comparison. rather than raw image detail information. Transposed convolution introduces in￾creased computation and latency. To address the features misalignment problem with minimal cost while maintaining the speed of near… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Illustrations of the ESD-I and ESD-II. It must be mentioned that although the size of the extended window does not have to be same in the ESD-I, pooling is a non-learnable operation, so the features from an unequal window may become noise for other and bring the troubl…
Figure 5
Figure 5. Figure 5: Illustrations of the GSConvE-I and II [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 12 canonical work pages

  1. [1]

    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

  2. [2]

    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

  3. [3]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Dai, X., Chen, Y., Yang, J., Zhang, P., Yuan, L., Zhang, L.: Dynamic detr: End- to-end object detection with dynamic attention. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 2988–2997 (2021)

    work page 2021

  4. [4]

    In: 2009 IEEE conference on computer vision and pattern recognition

    Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., Fei-Fei, L.: Imagenet: A large- scale hierarchical image database. In: 2009 IEEE conference on computer vision and pattern recognition. pp. 248–255. Ieee (2009)

    2009

  5. [5]

    Everingham, M., Eslami, S.A., Van Gool, L., Williams, C.K., Winn, J., Zisserman, A.:Thepascalvisualobjectclasseschallenge:Aretrospective.Internationaljournal of computer vision111, 98–136 (2015)

    2015

  6. [6]

    arXiv preprint arXiv:1701.06659 (2017)

    Fu, C.Y., Liu, W., Ranga, A., Tyagi, A., Berg, A.C.: Dssd: Deconvolutional single shot detector. arXiv preprint arXiv:1701.06659 (2017)

    arXiv 2017

  7. [7]

    arXiv preprint arXiv:2107.08430 (2021)

    Ge, Z., Liu, S., Wang, F., Li, Z., Sun, J.: Yolox: Exceeding yolo series in 2021. arXiv preprint arXiv:2107.08430 (2021)

    arXiv 2021

  8. [8]

    In: Proceedings of the IEEE international conference on computer vision

    Girshick, R.: Fast r-cnn. In: Proceedings of the IEEE international conference on computer vision. pp. 1440–1448 (2015)

    work page 2015

Show all 40 references
  1. [9]

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

    Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for ac- curate object detection and semantic segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 580–587 (2014)

    2014

  2. [10]

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

    Han, K., Wang, Y., Tian, Q., Guo, J., Xu, C., Xu, C.: Ghostnet: More features from cheap operations. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1580–1589 (2020)

    2020

  3. [11]

    IEEE transactions on pattern analysis and machine intelligence 37(9), 1904–1916 (2015)

    He, K., Zhang, X., Ren, S., Sun, J.: Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE transactions on pattern analysis and machine intelligence 37(9), 1904–1916 (2015)

    2015

  4. [12]

    He,K.,Zhang,X.,Ren,S.,Sun,J.:Deepresiduallearningforimagerecognition.In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 770–778 (2016)

    2016

  5. [13]

    arXiv preprint arXiv:1704.04861 (2017)

    Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., An- dreetto, M., Adam, H.: Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861 (2017)

    2017 arXiv

  6. [14]

    1-tensorrt, tensorflow edge tpu and openvino export and inference

    Jocher, G., Chaurasia, A., Stoken, A., Borovec, J., Kwon, Y., Fang, J., Michael, K., Montes, D., Nadar, J., Skalski, P., et al.: ultralytics/yolov5: v6. 1-tensorrt, tensorflow edge tpu and openvino export and inference. Zenodo (2022) Rethinking Features-Fused-Pyramid-Neck for ...

    2022

  7. [15]

    https : / / github

    Jocher, G., et al.: Yolo-ultralytics. https : / / github . com / ultralytics / ultralytics (2023)

    2023

  8. [16]

    arXiv preprint arXiv:2209.02976 (2022)

    Li, C., Li, L., Jiang, H., Weng, K., Geng, Y., Li, L., Ke, Z., Li, Q., Cheng, M., Nie, W., et al.: Yolov6: A single-stage object detection framework for industrial applications. arXiv preprint arXiv:2209.02976 (2022)

    2022 arXiv

  9. [17]

    In: Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition

    Li, F., Zhang, H., Liu, S., Guo, J., Ni, L.M., Zhang, L.: Dn-detr: Accelerate detr training by introducing query denoising. In: Proceedings of the IEEE/CVF con- ference on computer vision and pattern recognition. pp. 13619–13627 (2022)

    2022

  10. [18]

    Journal of Real-Time Image Processing 21(3), 62 (2024)

    Li,H.,Li,J.,Wei,H.,Liu,Z.,Zhan,Z.,Ren,Q.:Slim-neckbygsconv:alightweight- design for real-time detector architectures. Journal of Real-Time Image Processing 21(3), 62 (2024)

    2024

  11. [19]

    In: Proceedings of the IEEE/CVF international conference on computer vision

    Li, Y., Chen, Y., Wang, N., Zhang, Z.: Scale-aware trident networks for object detection. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 6054–6063 (2019)

    2019

  12. [20]

    arXiv preprint arXiv:1712.00960 (2017)

    Li, Z., Yang, L., Zhou, F.: Fssd: feature fusion single shot multibox detector. arXiv preprint arXiv:1712.00960 (2017)

    2017 arXiv

  13. [21]

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

    Lin, T.Y., Dollár, P., Girshick, R., He, K., Hariharan, B., Belongie, S.: Feature pyramid networks for object detection. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 2117–2125 (2017)

    2017

  14. [22]

    In: Computer Vision– ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13

    Lin, T.Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., Zitnick, C.L.: Microsoft coco: Common objects in context. In: Computer Vision– ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part V 13. pp. 740–755...

    2014

  15. [23]

    arxiv 2019

    Liu, S., Huang, D., Wang, Y.: Learning spatial fusion for single-shot object detec- tion. arxiv 2019. arXiv preprint arXiv:1911.09516 (1911)

    1911 arXiv

  16. [24]

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

    Liu, S., Qi, L., Qin, H., Shi, J., Jia, J.: Path aggregation network for instance segmentation. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 8759–8768 (2018)

    2018

  17. [25]

    In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14

    Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C.Y., Berg, A.C.: Ssd: Single shot multibox detector. In: Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14. pp. 21–37. Springer (2016)

    2016

  18. [26]

    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

  19. [27]

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

    Liu, Z., Mao, H., Wu, C.Y., Feichtenhofer, C., Darrell, T., Xie, S.: A convnet for the 2020s. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 11976–11986 (2022)

    2022

  20. [28]

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

    Qiao, S., Chen, L.C., Yuille, A.: Detectors: Detecting objects with recursive feature pyramid and switchable atrous convolution. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 10213–10224 (2021)

    2021

  21. [29]

    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)

    2016

  22. [30]

    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

  23. [31]

    arXiv preprint arXiv:1804.02767 (2018)

    Redmon, J., Farhadi, A.: Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767 (2018)

    2018 arXiv

  24. [32]

    arXiv preprint arXiv:1409.1556 (2014) 18 Hulin Li

    Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014) 18 Hulin Li

    2014 arXiv

  25. [33]

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

    Singh, B., Davis, L.S.: An analysis of scale invariance in object detection snip. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 3578–3587 (2018)

    2018

  26. [34]

    Tan,M.,Pang,R.,Le,Q.V.:Efficientdet:Scalableandefficientobjectdetection.In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 10781–10790 (2020)

    2020

  27. [35]

    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

  28. [36]

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

    Wu, Y., Chen, Y., Yuan, L., Liu, Z., Wang, L., Li, H., Fu, Y.: Rethinking classi- fication and localization for object detection. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 10186–10195 (2020)

    2020

  29. [37]

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

    Yang, Z., Zhou, Y., Chen, Z., Ngiam, J.: 3d-man: 3d multi-frame attention network for object detection. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 1863–1872 (2021)

    2021

  30. [38]

    arXiv preprint arXiv:2203.03605 (2022)

    Zhang, H., Li, F., Liu, S., Zhang, L., Su, H., Zhu, J., Ni, L.M., Shum, H.Y.: Dino: Detr with improved denoising anchor boxes for end-to-end object detection. arXiv preprint arXiv:2203.03605 (2022)

    2022 arXiv

  31. [39]

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

    Zhang, X., Zhou, X., Lin, M., Sun, J.: Shufflenet: An extremely efficient convolu- tional neural network for mobile devices. In: Proceedings of the IEEE conference on computer vision and pattern recognition. pp. 6848–6856 (2018)

    2018

  32. [40]

    arXiv preprint arXiv:2010.04159 (2020)

    Zhu, X., Su, W., Lu, L., Li, B., Wang, X., Dai, J.: Deformable detr: Deformable transformers for end-to-end object detection. arXiv preprint arXiv:2010.04159 (2020)

    2020 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.