Pith. sign in

REVIEW 3 major objections 3 minor 1 cited by

Asymmetric Non-local Neural Networks for Semantic Segmentation

T0 review · 3 major / 3 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read Replacing a non-local block's dense pairwise attention with 110 pyramid-pooled anchors keeps segmentation accuracy while cutting memory by 28x.

desk verdict Good efficiency engineering for non-local blocks, but the 'without sacrificing performance' claim is not backed by the reported experiments. read the letter →

arxiv 1908.07678 v5 pith:5K46PLPO submitted 2019-08-21 cs.CV cs.LG

Zhen Zhu , Mengde Xu , Song Bai , Tengteng Huang , Xiang Bai This is my paper · ORCID
classification cs.CVcs.LG
keywords semanticsegmentationnon-localneuralnetworksspatialpyramidpoolingasymmetricattentionCityscapesADE20KPASCALContextmulti-levelfeaturefusion
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 claims that the cost of non-local attention in semantic segmentation can be cut dramatically without losing accuracy by attending over a small set of sampled anchor points instead of every pixel. Its Asymmetric Pyramid Non-local Block (APNB) applies spatial pyramid average pooling to the key and value embeddings, shrinking the attention similarity matrix from $N\times N$ to $N\times S$ with $S=110$ anchors. The authors report that APNB matches or slightly beats a standard non-local block on Cityscapes validation ($78.6$ vs $78.4$ mIoU) while using about $28\times$ less GPU memory and running about $6\times$ faster at a $256\times 128$ input. A second block, AFNB, applies the same idea to fuse high- and low-level feature maps, and the full network reaches $81.3$ mIoU on the Cityscapes test set. The wider point is that long-range context for segmentation does not require dense pairwise comparison.

What carries the argument

The load-bearing mechanism is the asymmetric non-local block built on the observation that the output size of a non-local block is fixed once the query branch's $N$ and channel count are set; the key and value branches may carry any number $S$ of locations. APNB instantiates this by inserting a parameter-free spatial pyramid pooling module---adaptive average pools of sizes $1,3,6,8$, flattened and concatenated---after the key and value embeddings, producing $S=110$ anchor points. The same construction with two input feature maps gives AFNB, which fuses Stage4 and Stage5 features through the same $N\times S$ attention pattern. Sharing the $1\times 1$ convolutions for key and value further cuts parameters. The mechanism carrying the argument is therefore the replacement of the dense $N\times N$ similarity matrix by an $N\times S$ one, with $S$ chosen so that the pooled anchors carry global, multi-scale statistics.

What would settle it

Run the exact same training setup but replace the pyramid average pooling with a random selection of the same $110$ anchor locations from the key and value maps. If Cityscapes validation mIoU stays near $79.9$, the pyramid statistics are not what preserves accuracy; if it drops, the pyramid sampling is carrying the claim. A second check is to plot mIoU against anchor count $S\in\{50,110,225\}$: if accuracy keeps rising past $110$, the 'without sacrificing performance' claim is only partial.

Watch

Extended reading notes

Core claim

The central claim is that a non-local block's two large matrix multiplications, each $O(CH^2W^2)$, can be made asymmetric: only the query branch keeps all $N=H\cdot W$ locations, while the key and value branches are subsampled to $S$ representative points. Since the output shape of a non-local block is determined by the query and the channel dimension, changing $N$ to $S$ in the key/value branches leaves the output unchanged. The paper's specific instantiation is spatial pyramid average pooling with output sizes $\{1,3,6,8\}$, giving $S=110$ anchors, which reduces the multiplications to $O(CNS)$ and, at $H=128, W=256$, saves roughly $298\times$ of the multiplication cost. Empirically the approximation costs nothing: replacing a non-local block with APNB gives $78.6$ validation mIoU against $78.4$ for the non-local block, and the full AFNB+APNB model reaches $79.9$ on Cityscapes validation and $81.3$ on the test set. The ablation also shows that pyramid pooling outperforms max, random, and flat average sampling at equal anchor counts, and that accuracy rises with more anchors, which is why the paper attributes the preserved accuracy to the pyramid statistics rather than to the anchor count alone.

Load-bearing premise

The load-bearing premise is that the $S=110$ anchor points produced by pyramid average pooling over the key and value features preserve enough information for the attention operation, so replacing every pixel location with these anchors does not cost accuracy; the paper supports this only by experiments, with no error bound or theoretical guarantee.

Editorial extensions

If this is right

  • A non-local block can be made roughly $298\times$ cheaper in matrix-multiplication cost at a $256\times 128$ input by using $S=110$ anchors, with no loss of validation accuracy ($78.6$ vs $78.4$ mIoU on Cityscapes).
  • At the same input size APNB uses about $28\times$ less GPU memory ($277$ vs $7797$ MB) and runs about $6\times$ faster ($30.8$ vs $179.4$ ms) than the standard non-local block.
  • Long-range fusion of Stage4 and Stage5 features through AFNB improves over a common fusion module ($77.1$ vs $76.5$ mIoU), and the full model with both APNB and AFNB reaches $79.9$ mIoU on Cityscapes validation.
  • On the test set of Cityscapes the full network reports $81.3$ mIoU, on ADE20K validation $45.24$ mIoU, and on PASCAL Context validation $52.8$ mIoU, each above the other listed methods.

Reading between the lines

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

  • A testable extension is to apply the same asymmetric sampling to other attention-based tasks, since the output-shape argument only requires the query branch to keep all locations; the paper mentions future vision tasks but does not try them.
  • The anchor count $S=110$ was tuned on Cityscapes; on a 150-class dataset like ADE20K the curve in Table 7 might shift, so a grid over $S$ there would show whether the no-sacrifice claim generalizes.
  • The full model concatenates Stage5 features alongside AFNB's output, which the paper says protects training; ablating that skip connection would quantify how much of AFNB's benefit is real fusion versus stabilization.
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 / 3 minor

Summary. The manuscript proposes two non-local network variants for semantic segmentation: the Asymmetric Pyramid Non-local Block (APNB), which uses spatial pyramid pooling to sub-sample the key and value branches of the standard non-local block so that the attention matrix is N x S with S=110 instead of N x N, and the Asymmetric Fusion Non-local Block (AFNB), which adapts the same idea to fuse features from different stages. The network is built on a ResNet-101 FCN and evaluated on Cityscapes, ADE20K, and PASCAL Context, with additional results on COCO-Stuff-10K and NYUD-V2 in the appendix. The paper reports state-of-the-art or competitive mIoU scores and large efficiency gains: for a 256 x 128 input, APNB is about 6 times faster and uses about 28 times less GPU memory than a standard non-local block. Ablation studies on Cityscapes validation compare the proposed blocks with non-local and fusion baselines.

Significance. If the accuracy-parity claim holds, APNB is a genuinely useful contribution: it replaces the quadratic cost of non-local attention with a simple, parameter-free sampling module, and the efficiency gains are directly measured and backed by released code. The efficiency evidence is strong and concrete. However, the paper's central claim that APNB reduces computation 'without sacrificing performance' is not currently established: the only head-to-head evidence is a single-run 0.2 mIoU difference that is within the paper's own documented hardware/toolchain noise, and the APNB comparison is confounded by parameter sharing between the key and value branches. The AFNB/APNB combination is interesting, but the controlled evidence needs strengthening before the main claim can be accepted.

major comments (3)
  1. [Section 4.4, Table 6; Appendix C] The only evidence supporting the central claim that APNB does not sacrifice accuracy is the row pair '+ NB 78.4' vs '+ APNB 78.6'. Appendix C reports that the same model trained on 8x Titan V is about 0.5 mIoU better than when trained on 8x Titan Xp, and all ablation tables report single-run point estimates without variance. A 0.2 mIoU difference is therefore within the noise level the authors themselves document. To support the parity claim, the paper should report mean and standard deviation over multiple seeds (at least three) for the key rows, and should state the seeding protocol in the implementation details.
  2. [Section 3.4 and Figure 3; Section 3.1; Table 1] The comparison between APNB and the non-local block is confounded. Table 1 states that APNB and NB are 'entirely identical' apart from the pooling layers, but Section 3.4 says that in APNB the 1x1 convolutions W_theta and W_gamma share parameters, whereas the standard non-local block in Section 3.1 uses three separate 1x1 convolutions W_phi, W_theta, and W_gamma. The +APNB versus +NB comparison therefore changes two things at once: the pyramid sampling module and the weight-sharing scheme. The observed parity, or the small 0.2 gain, could be caused by the regularization effect of weight sharing alone. Please run and report NB with shared theta/gamma weights and APNB with separate weights, or otherwise isolate the effect of the pyramid sampling module.
  3. [Section 4.4, Tables 6 and 7] There is an apparent inconsistency in the ablation numbers. Table 7's default row (pyramid average, n=(1,3,6,8), S=110) reports 79.9 mIoU, which matches the Full model in Table 6, while Table 6's '+ APNB' row reports 78.6 mIoU for what is described as the same module. The text does not state whether the Table 7 experiments are run with APNB alone or with the full model including AFNB. This must be clarified, because Table 7 is used to select the pooling sizes and the sampling method; if Table 7 is the +APNB-alone configuration, the numbers are inconsistent, and if it is the full model, the hyperparameter selection was performed on the full system and should be reported as such.
minor comments (3)
  1. [Section 4.3.1] The text says 'The comparison results are given in Tab. 2' when referring to the single-block efficiency comparison; the actual table is Table 1, while Table 2 is the whole-network comparison.
  2. [Section 4.2] The implementation details report all results as point estimates. Please state the number of runs and the random seed policy used for each experiment, including the validation ablations.
  3. [Appendix C and Figures 5-6] The supplied PDF figures contain 'Huawei Confidential' watermarks; these should be removed for any public or camera-ready version of the paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the efficiency and accuracy claims are empirical benchmark results, not derivations from fitted inputs.

full rationale

The paper's derivation chain is constructive rather than circular. Section 3.2 introduces pyramid sampling after the key and value embeddings (Eqs. 8-10) and correctly counts the resulting complexity as O(CN S) instead of O(CN^2); Eq. (13), T = S/N, is a direct arithmetic consequence of the construction, not a fitted prediction. The claim that APNB is 'without sacrificing performance' is supported by the ablation in Table 6 (78.4 vs 78.6 mIoU) and Table 7, which are empirical measurements on the Cityscapes validation set, not quantities derived from the method's own assumptions. The sampling levels {1,3,6,8} are selected from ablation results, but this is standard hyperparameter selection on a validation benchmark rather than a parameter fitted to data and then renamed as a prediction; no equation in the paper reduces to its own input. The only self-involvement is the citation of the authors' own torchcv repository [37] as the implementation base for training, which is not load-bearing for any scientific claim and does not substitute for the external benchmarks. Appendix C's note that hardware can shift mIoU by about 0.5 points, and the key/value weight-sharing confound noted in Section 3.4, are validity concerns about the strength of the empirical comparison, not circularity: they do not make any derived claim equivalent to its inputs. Because the central results are benchmark measurements obtained from externally defined datasets and baselines, the appropriate finding is no significant circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The paper relies on non-local attention [33] and spatial pyramid pooling [47]; its central efficiency claim rests on the empirical adequacy of sampled anchors. No new physical entities are introduced.

free parameters (3)
  • Pyramid pooling output sizes n = {1, 3, 6, 8} with S=110
    Selected by ablation on Cityscapes validation in Table 7; larger anchor sets improve mIoU, reaching 80.1 for 225 anchors, so the default setting is an efficiency and accuracy trade-off.
  • Auxiliary loss weight lambda = 0.4
    Fixed at 0.4 following PSPNet [47]; part of the training recipe and not central to the APNB claim.
  • Initial learning rates per dataset = 0.01 for Cityscapes and PASCAL Context, 0.02 for ADE20K
    Standard per-dataset tuning described in Section 4.2; not load-bearing for the core contribution.
assumptions (4)
  • domain assumption Pyramid-sampled anchors preserve enough information for non-local attention.
    Central to the APNB efficiency claim; introduced in Section 3.2 through Equations 8 to 10 and validated only empirically.
  • standard math Softmax is an appropriate normalization for the similarity matrix.
    Used in Section 3.1 after Equation 3 as the standard self-attention normalization.
  • domain assumption Average pooling summarizes local regions better than max or random sampling for this task.
    Supported empirically in Table 7 and attributed to PSPNet [47]; used for anchor sampling.
  • domain assumption ImageNet-pretrained ResNet-101 features transfer to semantic segmentation.
    Used in Section 4.2 as standard practice in the literature.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Asymmetric Non-local Neural Networks for Semantic Segmentation." pith.science (2026). https://pith.science/paper/5K46PLPO

@misc{pith2026190807678,
  author       = {Pith},
  title        = {Pith review of: Asymmetric Non-local Neural Networks for Semantic Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/5K46PLPO}},
  note         = {Machine review of arXiv:1908.07678}
}
read the original abstract

The non-local module works as a particularly useful technique for semantic segmentation while criticized for its prohibitive computation and GPU memory occupation. In this paper, we present Asymmetric Non-local Neural Network to semantic segmentation, which has two prominent components: Asymmetric Pyramid Non-local Block (APNB) and Asymmetric Fusion Non-local Block (AFNB). APNB leverages a pyramid sampling module into the non-local block to largely reduce the computation and memory consumption without sacrificing the performance. AFNB is adapted from APNB to fuse the features of different levels under a sufficient consideration of long range dependencies and thus considerably improves the performance. Extensive experiments on semantic segmentation benchmarks demonstrate the effectiveness and efficiency of our work. In particular, we report the state-of-the-art performance of 81.3 mIoU on the Cityscapes test set. For a 256x128 input, APNB is around 6 times faster than a non-local block on GPU while 28 times smaller in GPU running memory occupation. Code is available at: https://github.com/MendelXu/ANN.git.

Figures

Figures reproduced from arXiv: 1908.07678 by the authors.

Figure 1
Figure 1. Architecture of a standard non-local block (a) and the asymmet [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. GPU time ( ≥ 1 ms) comparison of different operations between a generic non-local block and our APNB. The last bin denotes the sum of all the time costs. The size of the inputs for these two blocks is 256 × 128. Some recent studies [20, 33, 47] indicate that the perfor￾mance could be improved if making sufficient use of long range dependencies. However, models that solely rely on convolutions exhibit limited ability… view at source ↗
Figure 3
Figure 3. Overview of the proposed Asymmetric Non-local Neural Network. In our implementation, the [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Demonstration of the pyramid max or average sampling process. [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparisons with DeepLab-V3 [ [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Qualitative comparisons among Full model and other variants of our model. The [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: Feature visualization of different stages on ResNet-101. The [PITH_FULL_IMAGE:figures/full_fig_p012_7.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Deep High-Resolution Representation Learning for Visual Recognition

    cs.CV 2019-08 conditional novelty 6.0 of 10

    HRNet maintains high-resolution feature maps in parallel with low-resolution streams and repeatedly fuses them, improving accuracy on pose, segmentation, detection, and face alignment benchmarks.

Reference graph

Works this paper leans on

54 extracted references · 48 canonical work pages · cited by 1 Pith paper

  1. [1]

    Segnet: A deep convolutional encoder-decoder architecture for image segmentation

    Vijay Badrinarayanan, Alex Kendall, and Roberto Cipolla. Segnet: A deep convolutional encoder-decoder architecture for image segmentation. IEEE Trans. Pattern Anal. Mach. Intell., 39(12):2481–2495, 2017. 1

    work page 2017

  2. [2]

    A non-local algorithm for image denoising

    Antoni Buades, Bartomeu Coll, and Jean-Michel Morel. A non-local algorithm for image denoising. In Proc. CVPR, pages 60–65, 2005. 1

    work page 2005

  3. [3]

    Holger Caesar, Jasper R. R. Uijlings, and Vittorio Ferrari. Coco-stuff: Thing and stuff classes in context. In Proc. CVPR, pages 1209–1218, 2018. 5

    work page 2018

  4. [4]

    Fast, exact and multi-scale inference for semantic image segmentation with deep gaussian crfs

    Siddhartha Chandra and Iasonas Kokkinos. Fast, exact and multi-scale inference for semantic image segmentation with deep gaussian crfs. In Proc. ECCV, pages 402–418, 2016. 2

    work page 2016

  5. [5]

    Liang-Chieh Chen, George Papandreou, Iasonas Kokkinos, Kevin Murphy, and Alan L. Yuille. Semantic image segmen- tation with deep convolutional nets and fully connected crfs. In Proc. ICLR, 2015. 1, 2

    work page 2015

  6. [6]

    Liang-Chieh Chen, George Papandreou, Iasonas Kokkinos, Kevin Murphy, and Alan L. Yuille. Deeplab: Semantic im- age segmentation with deep convolutional nets, atrous con- volution, and fully connected crfs.IEEE Trans. Pattern Anal. Mach. Intell., 40(4):834–848, 2018. 2, 4, 7, 11

    work page 2018

  7. [7]

    Rethinking atrous convolution for semantic image segmentation

    Liang-Chieh Chen, George Papandreou, Florian Schroff, and Hartwig Adam. Rethinking atrous convolution for semantic image segmentation. CoRR, abs/1706.05587, 2017. 6, 7, 8

    arXiv 2017

  8. [8]

    Encoder-decoder with atrous separable convolution for semantic image segmentation

    Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. Encoder-decoder with atrous separable convolution for semantic image segmentation. In Proc. ECCV, pages 833–851, 2018. 2

    work page 2018

Show all 54 references
  1. [9]

    The cityscapes dataset for semantic urban scene understanding

    Marius Cordts, Mohamed Omran, Sebastian Ramos, Timo Rehfeld, Markus Enzweiler, Rodrigo Benenson, Uwe Franke, Stefan Roth, and Bernt Schiele. The cityscapes dataset for semantic urban scene understanding. In Proc. CVPR, pages 3213–3223, 2016. 2, 5, 11, 12

    2016

  2. [10]

    Imagenet: A large-scale hierarchical image database

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Fei-Fei Li. Imagenet: A large-scale hierarchical image database. In Proc. CVPR, pages 248–255, 2009. 6, 11, 12

    2009

  3. [11]

    Context contrasted feature and gated multi- scale aggregation for scene segmentation

    Henghui Ding, Xudong Jiang, Bing Shuai, Ai Qun Liu, and Gang Wang. Context contrasted feature and gated multi- scale aggregation for scene segmentation. In Proc. CVPR, pages 2393–2402, 2018. 7, 11

    2018

  4. [12]

    Spatial pyramid pooling in deep convolutional networks for visual recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Spatial pyramid pooling in deep convolutional networks for visual recognition. IEEE Trans. Pattern Anal. Mach. Intell., 37(9):1904–1916, 2015. 1, 2

    1904

  5. [13]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proc. CVPR, pages 770–778, 2016. 5

    2016

  6. [14]

    Tsung-Wei Ke, Jyh-Jing Hwang, Ziwei Liu, and Stella X. Yu. Adaptive affinity fields for semantic segmentation. In Proc. ECCV, pages 605–621, 2018. 7

    2018

  7. [15]

    Lafferty, Andrew McCallum, and Fernando C

    John D. Lafferty, Andrew McCallum, and Fernando C. N. Pereira. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In Proc. ICML, pages 282–289, 2001. 2

    2001

  8. [16]

    Be- yond bags of features: Spatial pyramid matching for recog- nizing natural scene categories

    Svetlana Lazebnik, Cordelia Schmid, and Jean Ponce. Be- yond bags of features: Spatial pyramid matching for recog- nizing natural scene categories. In Proc. CVPR, pages 2169– 2178, 2006. 1, 2, 4

    2006

  9. [17]

    Dynamic- structured semantic propagation network

    Xiaodan Liang, Hongfei Zhou, and Eric Xing. Dynamic- structured semantic propagation network. In Proc. CVPR, pages 752–761, 2018. 7

    2018

  10. [18]

    Guosheng Lin, Anton Milan, Chunhua Shen, and Ian D. Reid. Refinenet: Multi-path refinement networks for high- resolution semantic segmentation. In Proc. CVPR, pages 5168–5177, 2017. 1, 4, 7, 11

    2017

  11. [19]

    Guosheng Lin, Chunhua Shen, Anton van den Hengel, and Ian D. Reid. Efficient piecewise training of deep structured models for semantic segmentation. In Proc. CVPR, pages 3194–3203, 2016. 7, 11

    2016

  12. [20]

    Wei Liu, Andrew Rabinovich, and Alexander C. Berg. Parsenet: Looking wider to see better. CoRR, abs/1506.04579, 2015. 1, 2

    2015 arXiv

  13. [21]

    The role of context for object detection and se- mantic segmentation in the wild

    Roozbeh Mottaghi, Xianjie Chen, Xiaobai Liu, Nam-Gyu Cho, Seong-Whan Lee, Sanja Fidler, Raquel Urtasun, and Alan Yuille. The role of context for object detection and se- mantic segmentation in the wild. In Proc. CVPR, 2014. 2, 5

    2014

  14. [22]

    Learning deconvolution network for semantic segmentation

    Hyeonwoo Noh, Seunghoon Hong, and Bohyung Han. Learning deconvolution network for semantic segmentation. In Proc. ICCV, pages 1520–1528, 2015. 2

    2015

  15. [23]

    Automatic differentiation in pytorch

    Adam Paszke, Sam Gross, Soumith Chintala, Gregory Chanan, Edward Yang, Zachary DeVito, Zeming Lin, Al- ban Desmaison, Luca Antiga, and Adam Lerer. Automatic differentiation in pytorch. 2017. 6

    2017

  16. [24]

    Large kernel matters - improve semantic segmenta- tion by global convolutional network

    Chao Peng, Xiangyu Zhang, Gang Yu, Guiming Luo, and Jian Sun. Large kernel matters - improve semantic segmenta- tion by global convolutional network. In Proc. CVPR, pages 1743–1751, 2017. 2, 7

    2017

  17. [25]

    U-net: Convolutional networks for biomedical image segmentation

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image segmentation. In Proc. MICCAI, pages 234–241, 2015. 2

    2015

  18. [26]

    Fully convolutional networks for semantic segmentation

    Evan Shelhamer, Jonathan Long, and Trevor Darrell. Fully convolutional networks for semantic segmentation. IEEE Trans. Pattern Anal. Mach. Intell., 39(4):640–651, 2017. 1, 2, 4, 7

    2017

  19. [27]

    Deep semantic segmentation for auto- mated driving: Taxonomy, roadmap and challenges

    Mennatullah Siam, Sara Elkerdawy, Martin J ¨agersand, and Senthil Yogamani. Deep semantic segmentation for auto- mated driving: Taxonomy, roadmap and challenges. InProc. ITSC, pages 1–8, 2017. 1

    2017

  20. [28]

    A 9 comparative study of real-time semantic segmentation for autonomous driving

    Mennatullah Siam, Mostafa Gamal, Moemen Abdel-Razek, Senthil Yogamani, Martin J ¨agersand, and Hong Zhang. A 9 comparative study of real-time semantic segmentation for autonomous driving. In CVPR workshop, pages 587–597,

  21. [29]

    Indoor segmentation and support inference from RGBD images

    Nathan Silberman, Derek Hoiem, Pushmeet Kohli, and Rob Fergus. Indoor segmentation and support inference from RGBD images. In Proc. ECCV, pages 746–760, 2012. 5, 11

    2012

  22. [30]

    Gomez, Lukasz Kaiser, and Illia Polosukhin

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszko- reit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Proc. NeurIPS, pages 6000–6010, 2017. 3

    2017

  23. [31]

    Gaussian conditional random field network for semantic segmentation

    Raviteja Vemulapalli, Oncel Tuzel, Ming-Yu Liu, and Rama Chellappa. Gaussian conditional random field network for semantic segmentation. In Proc. CVPR, pages 3224–3233,

  24. [32]

    Cottrell

    Panqu Wang, Pengfei Chen, Ye Yuan, Ding Liu, Zehua Huang, Xiaodi Hou, and Garrison W. Cottrell. Understand- ing convolution for semantic segmentation. In Proc. WACV, pages 1451–1460, 2018. 7

    2018

  25. [33]

    Girshick, Abhinav Gupta, and Kaiming He

    Xiaolong Wang, Ross B. Girshick, Abhinav Gupta, and Kaiming He. Non-local neural networks. In Proc. CVPR, pages 7794–7803, 2018. 1, 2, 3, 4

    2018

  26. [34]

    Wider or deeper: Revisiting the resnet model for visual recognition

    Zifeng Wu, Chunhua Shen, and Anton van den Hengel. Wider or deeper: Revisiting the resnet model for visual recognition. CoRR, abs/1611.10080, 2016. 7

    2016 arXiv

  27. [35]

    Unified perceptual parsing for scene understand- ing

    Tete Xiao, Yingcheng Liu, Bolei Zhou, Yuning Jiang, and Jian Sun. Unified perceptual parsing for scene understand- ing. In Proc. ECCV, pages 432–448, 2018. 7

    2018

  28. [36]

    Denseaspp for semantic segmentation in street scenes

    Maoke Yang, Kun Yu, Chi Zhang, Zhiwei Li, and Kuiyuan Yang. Denseaspp for semantic segmentation in street scenes. In Proc. CVPR, pages 3684–3692, 2018. 2, 6, 7

    2018

  29. [37]

    Torchcv: A pytorch-based framework for deep learning in computer vision

    Ansheng You, Xiangtai Li, Zhen Zhu, and Yunhai Tong. Torchcv: A pytorch-based framework for deep learning in computer vision. https://github.com/donnyyou/ torchcv, 2019. 6

    2019

  30. [38]

    Bisenet: Bilateral segmenta- tion network for real-time semantic segmentation

    Changqian Yu, Jingbo Wang, Chao Peng, Changxin Gao, Gang Yu, and Nong Sang. Bisenet: Bilateral segmenta- tion network for real-time semantic segmentation. In Proc. ECCV, pages 334–349, 2018. 5, 7

    2018

  31. [39]

    Learning a discriminative feature network for semantic segmentation

    Changqian Yu, Jingbo Wang, Chao Peng, Changxin Gao, Gang Yu, and Nong Sang. Learning a discriminative feature network for semantic segmentation. In Proc. CVPR, pages 1857–1866, 2018. 7

    2018

  32. [40]

    Multi-scale context aggrega- tion by dilated convolutions

    Fisher Yu and Vladlen Koltun. Multi-scale context aggrega- tion by dilated convolutions. CoRR, abs/1511.07122, 2015. 1, 2

    2015 arXiv

  33. [41]

    Deep layer aggregation

    Fisher Yu, Dequan Wang, Evan Shelhamer, and Trevor Dar- rell. Deep layer aggregation. In Proc. CVPR, pages 2403– 2412, 2018. 4

    2018

  34. [42]

    Ocnet: Object context net- work for scene parsing

    Yuhui Yuan and Jingdong Wang. Ocnet: Object context net- work for scene parsing. CoRR, abs/1809.00916, 2018. 5

    2018 arXiv

  35. [43]

    Dana, Jianping Shi, Zhongyue Zhang, Xiaogang Wang, Ambrish Tyagi, and Amit Agrawal

    Hang Zhang, Kristin J. Dana, Jianping Shi, Zhongyue Zhang, Xiaogang Wang, Ambrish Tyagi, and Amit Agrawal. Context encoding for semantic segmentation. In Proc. CVPR, pages 7151–7160, 2018. 1, 2, 6, 7

    2018

  36. [44]

    Goodfellow, Dimitris N

    Han Zhang, Ian J. Goodfellow, Dimitris N. Metaxas, and Augustus Odena. Self-attention generative adversarial net- works. In Proc. ICML, pages 7354–7363, 2019. 1, 3

    2019

  37. [45]

    Scale-adaptive convolutions for scene pars- ing

    Rui Zhang, Sheng Tang, Yongdong Zhang, Jintao Li, and Shuicheng Yan. Scale-adaptive convolutions for scene pars- ing. In Proc. ICCV, pages 2050–2058, 2017. 7

    2017

  38. [46]

    Exfuse: Enhancing feature fusion for semantic segmentation

    Zhenli Zhang, Xiangyu Zhang, Chao Peng, Xiangyang Xue, and Jian Sun. Exfuse: Enhancing feature fusion for semantic segmentation. In Proc. CVPR, pages 273–288, 2018. 4, 11

    2018

  39. [47]

    Pyramid scene parsing network

    Hengshuang Zhao, Jianping Shi, Xiaojuan Qi, Xiaogang Wang, and Jiaya Jia. Pyramid scene parsing network. In Proc. CVPR, pages 6230–6239, 2017. 1, 2, 4, 5, 6, 7, 8

    2017

  40. [48]

    Psanet: Point- wise spatial attention network for scene parsing

    Hengshuang Zhao, Yi Zhang, Shu Liu, Jianping Shi, Chen Change Loy, Dahua Lin, and Jiaya Jia. Psanet: Point- wise spatial attention network for scene parsing. In Proc. ECCV, pages 270–286, 2018. 5, 6, 7

    2018

  41. [49]

    Shuai Zheng, Sadeep Jayasumana, Bernardino Romera- Paredes, Vibhav Vineet, Zhizhong Su, Dalong Du, Chang Huang, and Philip H. S. Torr. Conditional random fields as recurrent neural networks. In Proc. ICCV, pages 1529–1537,

  42. [50]

    Scene parsing through ADE20K dataset

    Bolei Zhou, Hang Zhao, Xavier Puig, Sanja Fidler, Adela Barriuso, and Antonio Torralba. Scene parsing through ADE20K dataset. In Proc. CVPR, pages 5122–5130, 2017. 2, 5

    2017

  43. [51]

    Similarity fusion for visual tracking

    Yu Zhou, Xiang Bai, Wenyu Liu, and Longin Jan Latecki. Similarity fusion for visual tracking. International Journal of Computer Vision, 118(3):337–363, 2016. 4

    2016

  44. [52]

    Prior-aware neu- ral network for partially-supervised multi-organ segmenta- tion

    Yuyin Zhou, Zhe Li, Song Bai, Chong Wang, Xinlei Chen, Mei Han, Elliot Fishman, and Alan Yuille. Prior-aware neu- ral network for partially-supervised multi-organ segmenta- tion. In Proc. ICCV, 2019. 1

    2019

  45. [53]

    Semi-supervised 3d ab- dominal multi-organ segmentation via deep multi-planar co- training

    Yuyin Zhou, Yan Wang, Peng Tang, Song Bai, Wei Shen, Elliot Fishman, and Alan Yuille. Semi-supervised 3d ab- dominal multi-organ segmentation via deep multi-planar co- training. In Proc. WACV, pages 121–140, 2019. 1 10 A. Quantitative comparisons on COCO-Stuff- 10K and NYUD-V2...

    2019

  46. [54]

    is a small benchmark and COCO-Stuff-10K is quite challenging and large, these results further verify the effec- tiveness of our method on both small and large benchmarks. Method BackbonemIoU (%)Method BackbonemIoU (%)RefineNet [18]ResNet-10133.6 Piecewise [19]VGG16 40.6CCL [11]...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.