Pith. sign in

REVIEW 3 major objections 5 minor 33 references

Dual Atrous Separable Convolution for Improving Agricultural Semantic Segmentation

T0 review · 3 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read A lightweight convolutional model beats a transformer baseline on agricultural field-anomaly segmentation.

desk verdict A plausible lightweight agricultural segmentation module with a useful ablation trail, but the headline comparison to prior work is not trustworthy because the evaluation protocol was changed. read the letter →

arxiv 2506.22570 v1 pith:AN36P6XF submitted 2025-06-27 cs.CV

Chee Mei Ling , Thangarajah Akilan , Aparna Ravinda Phalke This is my paper · ORCID
classification cs.CV
keywords semanticsegmentationprecisionagricultureatrousconvolutiondepthwiseseparableDeepLabV3Agriculture-Visionmodelefficiencyskipconnection
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 argues that a convolutional encoder-decoder can reach the accuracy of heavy vision-transformer models on agricultural anomaly segmentation while staying light enough for drones and edge devices. To do this it replaces the plain atrous convolutions in DeepLabV3's Atrous Spatial Pyramid Pooling with a Dual Atrous Separable Convolution (DAS-Conv) module, in which a standard dilated convolution and a depthwise-dilated-plus-pointwise convolution run in parallel and their outputs are concatenated. Combined with a skip connection from the first encoder block and dilation rates $\{4,8,12,24\}$, the model reports 47.17% mean IoU on the Agriculture-Vision validation set, 8.69 points above the Agri-Vision baseline, using 7.6M parameters and 6.32 GFLOPs. The practical stake is that a carefully tuned lightweight CNN can deliver transformer-like segmentation accuracy at a fraction of the compute.

What carries the argument

The load-bearing object is the DAS-Conv module, a dual-branch convolutional unit. One branch applies a standard $3\times3$ atrous convolution with dilation $d$; the other applies a depthwise atrous $3\times3$ convolution followed by a $1\times1$ pointwise convolution. Each branch reduces channels to one tenth of the input, and the two outputs are concatenated channel-wise to form the module output. The module is used in place of each atrous convolution in the ASPP sub-network, with dilation rates 4, 8, 12, and 24 and padding $\lceil d(K-1)/2\rceil$. A second mechanism is the skip connection from the encoder's first block, carrying $64\times64\times40$ feature maps into the decoder to re-inject low-level spatial detail before the final $8\times$ upsampling. Together these carry the argument that multiscale receptive-field coverage can be widened cheaply without the parameter cost of transformer attention.

What would settle it

Retrain the proposed model and the two transformer baselines under one identical pipeline, with the same augmentation, input channel order, loss, batch size, and validation split, and compare mIoU on Agriculture-Vision; if the proposed model no longer beats or matches the baselines, the paper's comparative claim collapses.

Watch

Extended reading notes

Core claim

The paper's central claim is that swapping standard atrous convolutions for the DAS-Conv module inside DeepLabV3's ASPP, and adding a skip connection from the $64\times64\times40$ stage of a MobileNetV3-Large encoder, produces a segmentation model that beats its baseline and competes with much larger transformer models on Agriculture-Vision. The model reports mIoU 47.17% on the validation set, against 43.40% for the Agri-Vision baseline and 46.50% for SegFormer, while SegFormer+MiT-B5 reaches 49.04%. It does this with 7.6M parameters and 6.32 GFLOPs, versus 84.7M parameters and 183.3 GFLOPs for SegFormer+MiT-B5. On the paper's effectiveness metric, mIoU gain divided by $\log(\text{params})\times\text{GFLOPs}$, the model scores 67.77% compared with 1.60% for SegFormer+MiT-B5, and it also reports per-class gains in Endrow, Planter Skip, and Weed Cluster.

Load-bearing premise

The central comparison assumes that published mIoU values from other studies, obtained under different training pipelines, augmentations, class-label flattening, and validation protocols, can be compared directly to this paper's validation number, so that the reported margin over SegFormer and the 67.77% effectiveness score are meaningful.

Editorial extensions

If this is right

  • The reported accuracy at 6.32 GFLOPs implies the same segmentation quality can run on drone- and edge-class hardware where 84.7M-parameter transformer models cannot.
  • The tuning result, dilation rates $\{4,8,12,24\}$ beating $\{12,24,32\}$ on $32\times32$ feature maps, suggests large dilation rates add redundancy on small feature maps.
  • The ablation path indicates that running a regular atrous branch and an atrous-separable branch in parallel helps more than either branch alone.
  • Four-channel nRGB input (RGB plus near-infrared) contributes to the gain over RGB-only input.
  • The skip connection from the $64\times64\times40$ encoder stage outperforms both a coarser $32\times32\times960$ and a finer $128\times128\times24$ skip source.

Reading between the lines

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

  • The efficiency metric used here weights parameters and FLOPs multiplicatively; if that metric were adopted more widely, it would reshape how segmentation models are compared, but that is a modelling choice rather than a fact about model quality.
  • The dual-branch atrous design could plausibly transfer to other ASPP-style decoders handling small irregular objects, such as medical lesion or urban damage segmentation; the paper tests only agriculture.
  • A sharper check of the paper's real-world premise would be to measure wall-clock latency and energy on the actual edge device, since GFLOPs do not fully determine speed on mobile hardware.
  • Because the Agriculture-Vision labels are non-exclusive and the paper flattens them by pixel frequency, the reported class IoUs depend on that flattening rule; a different priority order could shift per-class numbers.
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

3 major / 5 minor

Summary. The paper proposes Dual Atrous Separable Convolution (DAS-Conv), a module that combines a standard atrous convolution in parallel with a depthwise atrous separable convolution and concatenates their outputs, integrated into the ASPP module of DeepLabV3 with a MobileNetV3-Large encoder. A skip connection from the first encoder block to the decoder is also added. The model is trained and evaluated on the Agriculture-Vision v2 dataset with a non-standard preprocessing that flattens overlapping labels and excludes the storm-damage class. The authors report 47.17% mIoU on the validation set, an 8.69% relative improvement over the Agri-Vision baseline (43.40), with 7.6M parameters and 6.32 GFLOPs, and claim efficiency gains via a self-defined effectiveness metric. The paper includes a systematic ablation study in Appendix C and a PyTorch code snippet in Appendix A.

Significance. If the reported comparisons were protocol-matched, the contribution would be meaningful: a lightweight convolutional model rivaling much larger transformer-based segmenters on a real agricultural benchmark would be a practically useful result. The manuscript's strengths are the clear module formulation (Section 3.2.2 and Appendix A), the documented ablation path from a vanilla ResNet50 model to the final configuration (Appendix C), and the use of the same validation images as prior work for qualitative comparison. However, the central quantitative claims depend on comparing the authors' own evaluation protocol with published numbers obtained under different protocols, so the significance is currently conditional on re-establishing those comparisons fairly.

major comments (3)
  1. [Section 4.1.1 and Table 4] The preprocessing described in Section 4.1.1 changes the evaluation target relative to the published baselines cited in Table 4. The mIoU in Eq. (6) is computed against ground-truth maps that have been flattened by assigning overlapping labels according to pixel frequency, and the storm-damage class is excluded from the averaging set. The published numbers for the Agri-Vision baseline [5], AAFormer [21], and SegFormer [25] were obtained under their own protocols. Because removing a rare class from a mean can mechanically increase mIoU, and because flattening alters the ground truth itself, the reported 47.17% and the claimed 8.69% improvement over the baseline do not by themselves establish an architectural advantage. The authors should either re-run all baselines under exactly their own preprocessing and class set, or adopt the official Agriculture-Vision evaluation protocol for their own model, and then recompute every comparison in Tables 4 and 5.
  2. [Section 4.3, Eq. (7), Table 5] The 'effectiveness' metric in Eq. (7) is defined ad hoc, and comparing its values across models is not meaningful unless DiffmIoU is computed against the same baseline under the same protocol. The value 67.77% for the proposed model versus 1.60% for SegFormer+MiT-B5 therefore does not support the abstract's claim of 'more than 66% improvement in efficiency'. The metric also lacks a clear unit, the log(Params) term is arbitrary, and no standard efficiency measure such as latency, throughput, or accuracy-per-FLOP on fixed hardware is reported. The efficiency claim should be either removed or replaced with a protocol-matched, standard efficiency comparison.
  3. [Section 4.2 and Table 4] All reported mIoU values appear to come from a single training run; Section 4.2 lists one optimizer configuration and does not mention seeds or repeated runs. The margin over SegFormer in Table 4 is only 0.67 mIoU points (47.17 vs 46.50), and the margin over the Agri-Vision baseline is 3.77 points. Without variance estimates or a statistical comparison, the claim of 'outperforming' these models is fragile. The authors should report the mean and standard deviation over at least three independent training runs for the proposed model and, ideally, for the re-run baselines.
minor comments (5)
  1. [Section 4.4 and Table 5] The sentence 'our model is roughly 6× smaller than MiT-B3 and 11× smaller than MiT-B5' is correct in terms of parameter counts, but the GFLOPs for the baselines are cited from [9,30] while the authors' GFLOPs are self-reported. Please state the input resolution and the tool or convention used to measure GFLOPs for all models so the numbers are reproducible.
  2. [Abstract and Section 4.4] The phrase 'more than 66% improvement in efficiency' is unclear: Table 5 reports an effectiveness score of 67.77% for the proposed model and 1.60% for SegFormer+MiT-B5, which is a factor of roughly 42, not a percentage improvement of 66%. Please clarify whether the claim refers to the absolute score, a relative improvement, or something else.
  3. [Appendix C and Table 6] The notation '322 × 960' is ambiguous and should read '32^2 × 960' (similarly for other entries in Table 6 and Table 2), and the input row '5122 × 4' should read '512^2 × 4'.
  4. [References] References [29] and [30] are duplicates of the same SegFormer paper; please consolidate them into a single entry.
  5. [Throughout] The manuscript contains several typographical and presentation errors, including 'datset' (Appendix C), 'convlutional' (Impact Statement), 'UA Vs' (Section 1), a duplicated opening paragraph in Appendix A, duplicated text between Appendices A and B, and a repeated 'Double Plant' header in Figure 5's caption. These should be cleaned up.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular reasoning: the architecture choices are empirical ablations, and the headline comparison is threatened by an evaluation-protocol mismatch, not by a self-referential derivation.

full rationale

This is an empirical architecture paper with no first-principles derivation whose conclusion is fed back into its inputs. The proposed DAS-Conv design, dilation rates {4,8,12,24}, skip-connection placement, optimizer, and input channel order are selected through the ablation experiments in Appendix C and Table 6; selecting hyperparameters by validation performance is model search, not circularity. The self-defined effectiveness metric of Eq. (7) is explicitly stated and is not a 'prediction', although its Diff_mIoU term inherits the baseline-comparability caveat. Self-citations [1,12,24] appear only as general background in the introduction and are not load-bearing. The principal weakness is that Section 4.1.1 flattens overlapping Agriculture-Vision labels and excludes the scarce storm-damage class, while the published baseline and SOTA numbers in Table 4 (e.g., Agri-Vision baseline 43.40) were not re-run under that same protocol; consequently the reported 47.17 mIoU and the 8.69% gain may not be directly comparable. This is a validity and comparability concern, not a circularity: the reported numbers are not equal to their inputs by construction.

Assumptions & free parameters 2 free parameters · 3 assumptions · 0 invented entities

The central claims rest on the dataset, the cross-paper comparison protocol, and the single-run validation mIoU. The DAS-Conv module itself is a new architectural component, but no new physical or theoretical entities are introduced.

free parameters (2)
  • Dilation rate set {4, 8, 12, 24} = 4, 8, 12, 24
    Chosen via ablation on the validation set to maximize mIoU, as described in Stage 5 of Appendix C.
  • Skip connection source feature map 64x64x40 = 64x64x40
    Selected through ablation in Stage 4 of Appendix C after comparing 32x32x960, 64x64x40, and 128x128x24 feature maps.
assumptions (3)
  • domain assumption The Agriculture-Vision ground truth annotations are reliable, and the label-flattening priority based on pixel frequency is acceptable.
    The dataset is taken as ground truth without verification; the flattening rule is introduced in Section 4.1.1.
  • domain assumption Previously published mIoU numbers in Table 4 are accurate and directly comparable.
    The paper does not re-run prior models; it cites their reported values as if they share the same protocol, as seen in Table 4 and Section 4.4.
  • domain assumption Validation mIoU is an appropriate objective without uncertainty quantification.
    All model selection and claims are based on a single validation set metric, with no confidence intervals or statistical tests.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Dual Atrous Separable Convolution for Improving Agricultural Semantic Segmentation." pith.science (2026). https://pith.science/paper/AN36P6XF

@misc{pith2026250622570,
  author       = {Pith},
  title        = {Pith review of: Dual Atrous Separable Convolution for Improving Agricultural Semantic Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/AN36P6XF}},
  note         = {Machine review of arXiv:2506.22570}
}
read the original abstract

Agricultural image semantic segmentation is a pivotal component of modern agriculture, facilitating accurate visual data analysis to improve crop management, optimize resource utilization, and boost overall productivity. This study proposes an efficient image segmentation method for precision agriculture, focusing on accurately delineating farmland anomalies to support informed decision-making and proactive interventions. A novel Dual Atrous Separable Convolution (DAS Conv) module is integrated within the DeepLabV3-based segmentation framework. The DAS Conv module is meticulously designed to achieve an optimal balance between dilation rates and padding size, thereby enhancing model performance without compromising efficiency. The study also incorporates a strategic skip connection from an optimal stage in the encoder to the decoder to bolster the model's capacity to capture fine-grained spatial features. Despite its lower computational complexity, the proposed model outperforms its baseline and achieves performance comparable to highly complex transformer-based state-of-the-art (SOTA) models on the Agriculture Vision benchmark dataset. It achieves more than 66% improvement in efficiency when considering the trade-off between model complexity and performance, compared to the SOTA model. This study highlights an efficient and effective solution for improving semantic segmentation in remote sensing applications, offering a computationally lightweight model capable of high-quality performance in agricultural imagery.

Figures

Figures reproduced from arXiv: 2506.22570 by the authors.

Figure 1
Figure 1. The proposed framework. For details, refer to Table 2: encoder backbone, Table 3: ASPP, and Figure 2: [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. An illustration of the proposed DAS-Conv module. This study conducted extensive experiments to identify the optimal placement, evaluating its impact at various stages within the architecture. The results show that the best placement is between Block 1 of the encoder and the decoder, as shown in [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. View of receptive field coverage for different Conv types and operation sequences in separable and atrous [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Semantic class vs pixel distribution in the dataset. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: Additional qualitative results. The results of AAFormer (MiT-B3 w/t boundary) are adopted from [21]. [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: Qualitative samples illustrating the proposed model’s failure case compared to the ground truth. [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 29 canonical work pages

  1. [5]

    Agriculture-vision: A large aerial image database for agricultural pattern analysis

    Mang Tik Chiu, Xingqian Xu, Yunchao Wei, Zilong Huang, Alexander G Schwing, Robert Brunner, Hrant Khachatrian, Hovnatan Karapetyan, Ivan Dozier, Greg Rose, et al. Agriculture-vision: A large aerial image database for agricultural pattern analysis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 2828–2838, 2020

    work page 2020

  2. [21]

    Aaformer: a multi-modal transformer network for aerial agricultural images

    Yao Shen, Lei Wang, and Yue Jin. Aaformer: a multi-modal transformer network for aerial agricultural images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 1705–1711, 2022

    work page 2022

  3. [25]

    Augmentation invariance and adaptive sampling in semantic segmentation of agricultural aerial images

    Antonio Tavera, Edoardo Arnaudo, Carlo Masone, and Barbara Caputo. Augmentation invariance and adaptive sampling in semantic segmentation of agricultural aerial images. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1656–1665, 2022

    work page 2022

  4. [1]

    Self-Supervised Learning for Image Segmentation: A Comprehensive Survey

    Thangarajah Akilan, Nusrat Jahan, and Wandong Zhang. Self-supervised learning for image segmentation: A comprehensive survey. arXiv preprint arXiv:2505.13584, 2025

    work page Pith review arXiv 2025

  5. [2]

    Ai- based uav swarms for monitoring and disease identification of brassica plants using machine learning: A review

    Zain Anwar Ali, Dingnan Deng, Muhammad Kashif Shaikh, Raza Hasan, and Muhammad Aamir Khan. Ai- based uav swarms for monitoring and disease identification of brassica plants using machine learning: A review. Computer Systems Science & Engineering, 48(1), 2024

    work page 2024

  6. [3]

    Reinforcement learning for sustainable agriculture

    Jonathan Binas, Leonie Luginbuehl, and Yoshua Bengio. Reinforcement learning for sustainable agriculture. In ICML 2019 workshop climate change: how can AI help, 2019

    work page 2019

  7. [4]

    Rethinking atrous convolution for semantic image segmentation

    Liang-Chieh Chen, George Papandreou, Florian Schroff, and Hartwig Adam. Rethinking atrous convolution for semantic image segmentation. arxiv. arXiv preprint arXiv:1706.05587, 5, 2017. 10 Dual Atrous Separable Convolution for Improving Agricultural Semantic Segmentation 2025, J UNE

    arXiv 2017

  8. [6]

    An image is worth 16x16 words: Transformers for image recognition at scale

    Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby. An image is worth 16x16 words: Transformers for image recognition at scale. In International Conference on Learning Representations, 2021

    2021

Show all 33 references
  1. [7]

    An accurate semantic segmentation model for bean seedlings and weeds identification based on improved erfnet

    Haozhang Gao, Mingyang Qi, Baoxia Du, Shuang Yang, Han Li, Tete Wang, Wenyu Zhong, and You Tang. An accurate semantic segmentation model for bean seedlings and weeds identification based on improved erfnet. Scientific Reports, 14(1):12288, 2024

    2024

  2. [8]

    Computer vision in smart agriculture and precision farm- ing: Techniques and applications

    Sumaira Ghazal, Arslan Munir, and Waqar S Qureshi. Computer vision in smart agriculture and precision farm- ing: Techniques and applications. Artificial Intelligence in Agriculture, 2024

    2024

  3. [9]

    Jiaqi Gu, Hyoukjun Kwon, Dilin Wang, Wei Ye, Meng Li, Yu-Hsin Chen, Liangzhen Lai, Vikas Chandra, and David Z. Pan. Multi-scale high-resolution vision transformer for semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (C...

    2022

  4. [10]

    Global context vision trans- formers

    Ali Hatamizadeh, Hongxu Yin, Greg Heinrich, Jan Kautz, and Pavlo Molchanov. Global context vision trans- formers. In International Conference on Machine Learning, pages 12633–12646. PMLR, 2023

    2023

  5. [11]

    Le, and Hartwig Adam

    Andrew Howard, Mark Sandler, Grace Chu, Liang-Chieh Chen, Bo Chen, Mingxing Tan, Weijun Wang, Yukun Zhu, Ruoming Pang, Vijay Vasudevan, Quoc V . Le, and Hartwig Adam. Searching for mobilenetv3. InProceed- ings of the IEEE/CVF International Conference on Computer Vision (ICCV),...

    2019

  6. [12]

    Improved semi-supervised attention gan for seman- tic segmentation

    Nusrat Jahan, Thangarajah Akilan, and Thanh Minh Nguyen. Improved semi-supervised attention gan for seman- tic segmentation. In 2024 IEEE Pacific Rim Conference on Communications, Computers and Signal Processing (PACRIM), pages 1–6. IEEE, 2024

    2024

  7. [13]

    Geminifusion: Effi- cient pixel-wise multimodal fusion for vision transformer

    Ding Jia, Jianyuan Guo, Kai Han, Han Wu, Chao Zhang, Chang Xu, and Xinghao Chen. Geminifusion: Effi- cient pixel-wise multimodal fusion for vision transformer. In Forty-first International Conference on Machine Learning, 2024

    2024

  8. [14]

    Segmentation of farmlands in aerial images by deep learning framework with feature fusion and context aggregation modules

    Sultan Daud Khan, Louai Alarabi, and Saleh Basalamah. Segmentation of farmlands in aerial images by deep learning framework with feature fusion and context aggregation modules. Multimedia Tools and Applications, 82(27):42353–42372, 2023

    2023

  9. [15]

    Multi-modal land cover mapping of remote sensing images using pyramid attention and gated fusion networks

    Qinghui Liu, Michael Kampffmeyer, Robert Jenssen, and Arnt-Børre Salberg. Multi-modal land cover mapping of remote sensing images using pyramid attention and gated fusion networks. International Journal of Remote Sensing, 43(9):3509–3535, 2022

    2022

  10. [16]

    Semantic segmentation of agricultural images: A survey

    Zifei Luo, Wenzhu Yang, Yunfeng Yuan, Ruru Gou, and Xiaonan Li. Semantic segmentation of agricultural images: A survey. Information Processing in Agriculture, 2023

    2023

  11. [17]

    A convolutional neural network ap- proach for image-based anomaly detection in smart agriculture

    Jos ´e Mendoza-Bernal, Aurora Gonz ´alez-Vidal, and Antonio F Skarmeta. A convolutional neural network ap- proach for image-based anomaly detection in smart agriculture. Expert Systems with Applications, 247:123210, 2024

    2024

  12. [18]

    How can precision farming work on a small scale? a systematic literature review

    Tam ´as Mizik. How can precision farming work on a small scale? a systematic literature review. Precision agriculture, 24(1):384–406, 2023

    2023

  13. [19]

    A novel hybrid methodology integrating pixel-and object-based techniques for mapping land use and land cover from high-resolution satellite data

    Muhammed Yusuf Ozturk and Ismail Colkesen. A novel hybrid methodology integrating pixel-and object-based techniques for mapping land use and land cover from high-resolution satellite data. International Journal of Remote Sensing, 45(16):5640–5678, 2024

    2024

  14. [20]

    U-net: Convolutional networks for biomedical image seg- mentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image seg- mentation. In Medical image computing and computer-assisted intervention–MICCAI 2015: 18th international conference, Munich, Germany, October 5-9, 2015, proceedings, par...

    2015

  15. [22]

    Effective data fusion with generalized vegetation index: Evidence from land cover segmentation in agriculture

    Hao Sheng, Xiao Chen, Jingyi Su, Ram Rajagopal, and Andrew Ng. Effective data fusion with generalized vegetation index: Evidence from land cover segmentation in agriculture. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, pages 6...

    2020

  16. [23]

    Nawaz, Asghar Ali Shah, Saim Rasheed, Sheeba Ilyas, and Muhammad Khurram Ehsan

    Ali Sohail, Naeem A. Nawaz, Asghar Ali Shah, Saim Rasheed, Sheeba Ilyas, and Muhammad Khurram Ehsan. A systematic literature review on machine learning and deep learning methods for semantic segmentation. IEEE Access, 10:134557–134570, 2022

    2022

  17. [24]

    Ecaseg: Enhancing semantic segmentation with edge context and attention strategy

    Tharrengini Suresh, Harish Sundaralingam, Thangarajah Akilan, and Nusrat Jahan. Ecaseg: Enhancing semantic segmentation with edge context and attention strategy. Procedia Computer Science, 260:274–282, 2025

    2025

  18. [26]

    Reslmffnet: a real-time semantic segmentation network for precision agriculture

    Irem Ulku. Reslmffnet: a real-time semantic segmentation network for precision agriculture. Journal of Real- Time Image Processing, 21(4):101, 2024

    2024

  19. [27]

    Attention is all you need

    A Vaswani. Attention is all you need. Advances in Neural Information Processing Systems, 2017

    2017

  20. [28]

    Np-semiseg: when neural processes meet semi-supervised semantic segmentation

    Jianfeng Wang, Daniela Massiceti, Xiaolin Hu, Vladimir Pavlovic, and Thomas Lukasiewicz. Np-semiseg: when neural processes meet semi-supervised semantic segmentation. In International Conference on Machine Learning, pages 36138–36156. PMLR, 2023

    2023

  21. [29]

    Segformer: Simple and efficient design for semantic segmentation with transformers

    Enze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar, Jose M Alvarez, and Ping Luo. Segformer: Simple and efficient design for semantic segmentation with transformers. Advances in neural information processing systems, 34:12077–12090, 2021

    2021

  22. [30]

    Alvarez, and Ping Luo

    Enze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, and Ping Luo. Segformer: Simple and efficient design for semantic segmentation with transformers. In M. Ranzato, A. Beygelzimer, Y . Dauphin, P.S. Liang, and J. Wortman Vaughan, editors, Advances in Neural I...

    2021

  23. [31]

    Reducing the feature divergence of rgb and near- infrared images using switchable normalization

    Siwei Yang, Shaozuo Yu, Bingchen Zhao, and Yin Wang. Reducing the feature divergence of rgb and near- infrared images using switchable normalization. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition workshops, pages 46–47, 2020

    2020

  24. [32]

    Netadapt: Platform-aware neural network adaptation for mobile applications

    Tien-Ju Yang, Andrew Howard, Bo Chen, Xiao Zhang, Alec Go, Mark Sandler, Vivienne Sze, and Hartwig Adam. Netadapt: Platform-aware neural network adaptation for mobile applications. In Proceedings of the European conference on computer vision (ECCV), pages 285–300, 2018

    2018

  25. [33]

    Agriculture-vision chal- lenge 2022–the runner-up solution for agricultural pattern recognition via transformer-based models

    Zhicheng Yang, Jui-Hsin Lai, Jun Zhou, Hang Zhou, Chen Du, and Zhongcheng Lai. Agriculture-vision chal- lenge 2022–the runner-up solution for agricultural pattern recognition via transformer-based models. arXiv preprint arXiv:2206.11920, 2022. 12 Dual Atrous Separable Convolut...

    2022 arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.