Pith. sign in

REVIEW 5 major objections 6 minor 55 references

Adversarially Domain-adaptive Latent Diffusion for Unsupervised Semantic Segmentation

T0 review · 5 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read ICCLD claims that a latent diffusion segmentation model with inter-coder skip connections and adversarial denoising achieves state-of-the-art unsupervised domain adaptation, with mIoU 74.4 on GTA5→Cityscapes and 67.2 on Synthia→Cityscapes.

desk verdict Plausible SOTA numbers undermined by a circular diffusion loss and internal inconsistencies; not referee-ready. read the letter →

arxiv 2412.16859 v2 pith:JSEIV4YM submitted 2024-12-22 cs.CV cs.AI

Jongmin Yu , Zhongtian Sun , Chen Bene Chi , Jinhong Yang , Shan Luo This is my paper · ORCID
classification cs.CVcs.AI
keywords unsuperviseddomainadaptationsemanticsegmentationlatentdiffusionmodelsadversariallearninginter-coderconnectionsdenoisingUNetGTA5toCityscapesSynthia
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 proposes Inter-Coder Connected Latent Diffusion (ICCLD), a latent diffusion segmentation model that adds long skip connections between its encoder and decoder and trains its denoising network with an adversarial domain-alignment loss. The goal is to transfer semantic segmentation from labeled synthetic images to unlabeled real images, avoiding expensive manual pixel annotations. The paper reports state-of-the-art results on the two standard benchmarks: 74.4 mIoU for GTA5→Cityscapes and 67.2 mIoU for Synthia→Cityscapes, surpassing prior methods by 0.6 and 1.4 mIoU. A two-stage recipe is central: first adapt the encoder and decoder on segmentation with mixed source-target images, then align latent distributions by training the denoiser to both predict noise and confuse a domain discriminator. If correct, the method shows that diffusion dynamics themselves can serve as a domain-alignment mechanism rather than only as a generative prior.

What carries the argument

The load-bearing mechanism is ICCLD, an LDM-style encoder-denoise-decoder stack. Its distinctive structural piece is the inter-coder connection: long skip connections that feed the encoder's multi-scale features directly into corresponding decoder blocks, preserving fine spatial detail that a latent bottleneck would lose. Its distinctive training piece is the second phase, where the denoising UNet ϵθ is conditioned on the clean latent z0 (Eq. 6) and optimized with a DDPM noise-prediction loss plus an adversarial KL-divergence loss (Eq. 7); the discriminator fdis decides whether a predicted noise came from the source, target, or mixed domain, and aligning those predicted-noise distributions is what the paper credits for closing the domain gap.

What would settle it

Run the full GTA5→Cityscapes pipeline with the noise-prediction loss of Equation (6) inactivated while keeping Equation (7); if mIoU stays at 74.4, the diffusion objective is not load-bearing. Alternatively, on the validation set, compute the correlation between the sampled noise ε and the predicted noise εθ: near-zero correlation would show the denoiser is not actually solving the stated denoising task.

Watch

Extended reading notes

Core claim

The paper's central claim is that inter-coder connected latent diffusion plus adversarial alignment sets a new state of the art for synthetic-to-real semantic segmentation without target labels. Training is split into two phases: first, the encoder and decoder are domain-adapted on segmentation using ClassMix-generated mixed images and pseudo-labels; second, with the encoder frozen, the denoising UNet is trained to predict the diffusion noise and to fool a discriminator that classifies predicted noise as source, target, or mixed. Across the two benchmarks the method reaches 74.4 mIoU for GTA5→Cityscapes and 67.2 mIoU for Synthia→Cityscapes, outperforming the previously published best results from DAFormer and HRDA. The ablation data support both design choices: the inter-coder connection improves boundary-level accuracy, and the adversarial second phase reduces false positives and produces cleaner masks.

Load-bearing premise

The approach depends on the assumption that conditioning the denoising network on the same clean latent used to build its noisy input leaves a genuine denoising task rather than an identity shortcut, so the second-phase gains can be attributed to diffusion-based alignment.

Editorial extensions

If this is right

  • Adding the adversarial second phase moves Step-1 performance from 68.4 to 74.4 mIoU on GTA5→Cityscapes and from 60.7 to 67.2 mIoU on Synthia→Cityscapes when the inter-coder connection is present.
  • The inter-coder connection alone lifts Step-1 performance from 58.3 to 68.4 mIoU on GTA5→Cityscapes and from 42.7 to 60.7 mIoU on Synthia→Cityscapes.
  • A diffusion-based UDA model can outperform non-diffusion state-of-the-art methods on both standard benchmarks, setting reference scores of 74.4 and 67.2 mIoU.
  • Class-wise, the method wins 11 of 19 classes on GTA5→Cityscapes and 12 of 16 on Synthia→Cityscapes, with the largest improvements concentrated in boundary-heavy classes such as Wall, Fence, and Rider.

Reading between the lines

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

  • A direct test of the mechanism would be to train the second phase with Equation (6) removed and only the adversarial loss active; comparable mIoU would indicate the diffusion objective is not the source of the gains.
  • Because the reported margins over HRDA are 0.6 and 1.4 mIoU, a repeated-seed study with standard deviations would determine whether the method reliably surpasses the prior state of the art.
  • The three-domain discriminator over predicted noise is portable: the same adversarial alignment could be applied to any feature-space UDA pipeline without requiring a diffusion model around it.
  • If the conditioning shortcut is absent in this architecture, the recipe suggests broad use of latent diffusion in dense prediction; if the shortcut is present, other conditional diffusion segmentation models that condition on the same clean latent would need re-examination.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The manuscript proposes ICCLD, a latent diffusion model with long inter-coder skip connections and adversarial alignment of the denoising UNet, for unsupervised domain adaptation in semantic segmentation. Training is done in two steps: first a teacher-student segmentation adaptation on the encoder/decoder, then adversarial domain alignment of the denoising network. The paper reports state-of-the-art mIoU scores of 74.4 on GTA5-to-Cityscapes and 67.2 on Synthia-to-Cityscapes, and presents ablations attributing large gains to the second phase, especially when the inter-coder connection is used.

Significance. If the proposed mechanism were sound, the paper would address a relevant problem and the two-step teacher-student diffusion design would be an interesting direction. The authors are to be credited for tackling difficult benchmarks and for reporting class-wise results. However, the central diffusion objective in Eq. (6) is degenerate as written, the adversarial loss in Eq. (7) is not a well-defined adversarial game, and the ablation numbers in the text disagree with Table 1. These issues prevent the paper from substantiating its main claim, and the reported SOTA margins over HRDA are small enough that repeated-seed evaluation would be needed even if the formulation were corrected.

major comments (5)
  1. [Section 3.4, Eq. (6)] The diffusion loss in Eq. (6) is degenerate as written. The noisy input is constructed as sqrt(alpha_bar_t) z* + sqrt(1 - alpha_bar_t) epsilon, and the conditioning input is z*_0, which the text identifies as the clean latent extracted from the same source, mixed, or target image. If z* and z*_0 are the same latent, the target noise is a deterministic function of the two inputs, epsilon = (x_t - sqrt(alpha_bar_t) z*_0) / sqrt(1 - alpha_bar_t), so any network can drive L_ldm to zero by implementing this arithmetic; the reverse process then returns the conditioning latent, making the diffusion path an identity. If z* and z*_0 are intended to be different, the distinction is never defined. This collapse is not discussed and undermines the attribution of the Step-2 gains in Table 1 to diffusion-based alignment.
  2. [Section 3.4, Eq. (7)] Equation (7) is not a coherent adversarial objective. The first expectation E[o* log fdis(...)] rewards the discriminator for predicting the true domain, while the KL-to-uniform term rewards a uniform prediction; optimizing the sum drives the two terms in opposite directions. No discriminator classification loss or alternating min-max update is defined, so it is not clear how fdis is trained or why this procedure aligns source and target latents. The paper also does not state the relative weight of L_adv in the total loss, although Table 1 attributes all Step-2 gains to this term.
  3. [Section 5.1 and Table 1] The ablation numbers in the text do not match Table 1. Section 5.1 states that the model with the inter-coder connection achieves 69.3 mIoU on Synthia-to-Cityscapes, but Table 1 lists 67.2 for Step 1&2 with the connection and 69.3 for Step 1&2 without the connection on GTA5-to-Cityscapes. Section 5.2 reports improvements of 9.0 and 3.8 mIoU from the second phase without the inter-coder connection, whereas Table 1 implies 11.0 and 15.8 mIoU (58.3 to 69.3 and 42.7 to 58.5). These inconsistencies make the central ablation impossible to verify.
  4. [Section 6, Table 2] The SOTA claim is not statistically supported. The reported margins over HRDA are +0.6 mIoU on GTA5-to-Cityscapes and +1.4 mIoU on Synthia-to-Cityscapes, yet the paper gives no number of independent runs, no standard deviations, and no significance test. In UDA semantic segmentation, margins of this size are often within run-to-run variability, so the claim that ICCLD outperforms existing methods is not established without repeated-seed evaluation.
  5. [Figure 1(b) and Section 3.3] The architecture description is incomplete. Figure 1(b) contains a block labelled 'Mask Encoder c Conditioning' that is never defined in the text, and Section 3.3 does not describe how the conditioning input z*_0 is injected into the denoising UNet (concatenation, cross-attention, or otherwise). Without this information the proposed mechanism cannot be reproduced or compared against alternative explanations of the reported gains.
minor comments (6)
  1. [Title, Abstract, Section 7, Figure 2] The method name is inconsistent: ICCLD in the title and abstract, 'Conditional and Inter-coder Connected Latent Diffusion (CICLD)' in Section 7, and 'Conditional and Long skip-connected Diffusion Model (CLDM)' in Figure 2.
  2. [Section 3.4, Step 1] The phrase 'prepared target label ys' appears to be a typo for the source label; as written it is unclear which labels are used for the mixed image.
  3. [Section 6, Table 2] The sentence claiming that class-wise performance differences are 'generally small—often below 0.2' is contradicted by Table 2, where several differences exceed 1 mIoU (e.g., Traffic Sign and Terrain on GTA5-to-Cityscapes).
  4. [Equation (6)] The notation in Eq. (6) is ambiguous: z* is used both as a generic latent variable and as a placeholder for zs, zs+t, zt, while z*_0 is defined only verbally; a precise index convention would help.
  5. [Section 4, Implementation details] Implementation details are incomplete: the paper does not specify the number of diffusion timesteps T used during training, the loss weight for L_adv, or the exact architecture of the discriminator output head, and no code is provided.
  6. [Figure 4] The qualitative comparison refers to 'green-coloured dotted boxes', but the figure as printed does not clearly show these boxes; please adjust the figure or the caption.

Circularity Check

1 steps flagged · score 8.0 of 10

Step-2 diffusion objective is degenerate: Eq. (6) conditions on the same clean latent used to build the noisy input, making the predicted noise a deterministic function of the inputs.

  1. self definitional [Section 3.4, Equation (6) and accompanying text (also Figure 2 caption)]
    "Since semantic segmentation is not just generating arbitrary segmentation masks. ICCLD should generate a suitable segmentation mask corresponding with the given image so that conditioning using the latent features is essential. ... the clear latent feature z0, extracted by E, is applied as a conditional factor during diffusion. The loss function for the t-step diffusion and de-noising processes is formulated as follows: Lldm(t, z∗, z∗0, ϵ) = ||ϵ − ϵθ(√¯αtz∗ + √1 − ¯αtϵ, t|z∗0)||^2, (6)"

    The noised input in Eq. (6) is √¯α_t z* + √(1−¯α_t)ϵ, and the conditioning input is z*0, which the text identifies as the clean latent extracted by the same encoder from the same images. On the only interpretation offered (z* = z*0, as in Figure 2 where z_t^0 is extracted from the input image and used as the condition), the target noise is exactly ϵ = (x_t − √¯α_t z*0)/√(1−¯α_t), a deterministic function of the two arguments supplied to ϵθ. The network can drive L_ldm to zero by learning this algebraic inversion, so the 'denoising prediction' is not an independent generative quantity. The Table 1 ablation that credits Step 2 with +8.0 and +6.5 mIoU therefore cannot be attributed to the stated diffusion objective; the diffusion path reduces to approximately identity at inference.

full rationale

The central load-bearing step is the second training phase: Table 1 attributes gains of 8.0 and 6.5 mIoU (GTA5 and Synthia to Cityscapes) to adversarial domain adaptation on the denoising UNet. This attribution depends on the diffusion loss in Eq. (6), but that loss is degenerate by construction: the noised latent is generated from the clean latent z*0 and the same z*0 is supplied as the conditioning input. The target ϵ is then recoverable by arithmetic from the two network inputs, so the model can minimize the loss without learning any denoising or generative behavior. At inference the iterated denoiser returns approximately the encoder's own latent, making the diffusion stage an identity-like fit rather than a distribution-aligning process. The adversarial loss in Eq. (7) is also under-specified: it combines a cross-entropy term with a KL-to-uniform term that pull fdis in opposite directions, and no discriminator loss or alternating min-max update is defined. These are internal derivational gaps rather than external-benchmark disagreements; the empirical mIoU numbers, even if reproducible, are not explained by the claimed diffusion-based alignment mechanism. No load-bearing self-citation pattern was found; the circularity is centered in the paper's own equations.

Assumptions & free parameters 5 free parameters · 5 assumptions · 2 invented entities

The central claim depends on a set of hand-chosen hyperparameters, on standard DDPM/LDM background, on the reliability of ClassMix pseudo-labels, and on two ad hoc modeling choices: the circular diffusion conditioning and the undefined mask encoder. The invented entities (inter-coder connection and mask encoder c) have no independent empirical or theoretical grounding beyond the paper's own ablations.

free parameters (5)
  • Adversarial loss weight
    Not specified in the paper; the balance between the diffusion loss and adversarial loss in the total objective is unknown, directly affecting the claimed contribution of adversarial learning.
  • EMA weight alpha = 0.999
    Hand-chosen update rate for the student model; standard but required to reproduce results.
  • Inference denoising steps = 50
    Number of DDIM steps at inference; affects runtime and output quality.
  • Initial learning rate = 6e-5
    Hand-chosen with a polynomial decay schedule; part of optimization settings that influence final mIoU.
  • Batch size and epochs = 2, 50
    Training settings that can affect convergence and final performance.
assumptions (5)
  • standard math Standard DDPM forward and reverse processes (Eqs. 1-4) are valid.
    Background from Ho et al. [11]; accepted.
  • standard math LDM backbone as defined by Rombach et al. [26] is used.
    The method relies on the pre-existing LDM architecture and latent space.
  • domain assumption ClassMix mixed images and pseudo-labels from the student are useful supervision for the target domain.
    Step 1 relies on this to adapt E and D; no validation of pseudo-label quality is provided.
  • ad hoc to paper Conditioning the denoising network on the clean latent does not trivialize the diffusion task.
    The paper provides no argument that the network does not simply copy the condition; the setup appears circular.
  • ad hoc to paper The discriminator with a uniform KL target produces domain-aligned latent features.
    Eq. 7 is introduced without a minimax derivation or ablation of the objective.
invented entities (2)
  • Inter-coder connection
    purpose: Long skip connections between encoder E and decoder D to preserve fine details.
    No theoretical or empirical evidence beyond the reported ablations; no formal definition of the connection in text.
  • Mask encoder c
    purpose: Appears in Figure 1 as a conditioning pathway feeding the denoising UNet.
    Never defined or described in the text; the conditioning mechanism is ambiguous.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Adversarially Domain-adaptive Latent Diffusion for Unsupervised Semantic Segmentation." pith.science (2026). https://pith.science/paper/JSEIV4YM

@misc{pith2026241216859,
  author       = {Pith},
  title        = {Pith review of: Adversarially Domain-adaptive Latent Diffusion for Unsupervised Semantic Segmentation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/JSEIV4YM}},
  note         = {Machine review of arXiv:2412.16859}
}
abstract

Semantic segmentation requires extensive pixel-level annotation, motivating unsupervised domain adaptation (UDA) to transfer knowledge from labelled source domains to unlabelled or weakly labelled target domains. One of the most efficient strategies involves using synthetic datasets generated within controlled virtual environments, such as video games or traffic simulators, which can automatically generate pixel-level annotations. However, even when such datasets are available, learning a well-generalised representation that captures both domains remains challenging, owing to probabilistic and geometric discrepancies between the virtual world and real-world imagery. This work introduces a semantic segmentation method based on latent diffusion models, termed Inter-Coder Connected Latent Diffusion (ICCLD), alongside an unsupervised domain adaptation approach. The model employs an inter-coder connection to enhance contextual understanding and preserve fine details, while adversarial learning aligns latent feature distributions across domains during the latent diffusion process. Experiments on GTA5, Synthia, and Cityscapes demonstrate that ICCLD outperforms state-of-the-art UDA methods, achieving mIoU scores of 74.4 (GTA5$\rightarrow$Cityscapes) and 67.2 (Synthia$\rightarrow$Cityscapes).

Figures

Figures reproduced from arXiv: 2412.16859 by the authors.

Figure 1
Figure 1. Illustration of the workflow for the two-step domain adaptation process using the proposed Inter-Coder Connected Latent [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Architectural details of ICCLD. For prediction a segmen [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Example images and labels of the (a) GTA-5 [ [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Qualitative comparison of the UDA performance of ICCLD according to the inter-coder connection and the usage of the extra DA [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Qualitative comparison on the UDA performance using the proposed ICCLD with HRDA [ [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

55 extracted references · 40 canonical work pages

  1. [1]

    Segdiff: Image segmentation with diffusion probabilistic mod- els

    Tomer Amit, Tal Shaharbany, Eliya Nachmani, and Lior Wolf. Segdiff: Image segmentation with diffusion probabilistic mod- els. arXiv preprint arXiv:2112.00390, 2021. 2, 3

    arXiv 2021

  2. [2]

    Label-efficient seman- tic segmentation with diffusion models

    Dmitry Baranchuk, Ivan Rubachev, Andrey V oynov, Valentin Khrulkov, and Artem Babenko. Label-efficient seman- tic segmentation with diffusion models. arXiv preprint arXiv:2112.03126, 2021. 2, 3

    arXiv 2021

  3. [3]

    Transunet: Transformers make strong encoders for medical image segmentation

    Jieneng Chen, Yongyi Lu, Qihang Yu, Xiangde Luo, Ehsan Adeli, Yan Wang, Le Lu, Alan L Yuille, and Yuyin Zhou. Transunet: Transformers make strong encoders for medical image segmentation. arXiv preprint arXiv:2102.04306, 2021. 2

    arXiv 2021

  4. [4]

    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 Proceedings of the IEEE conference on computer vision and pattern recognition, pages 3213–3223, 2016. 1, 5, 6

    work page 2016

  5. [5]

    Boxsup: Exploit- ing bounding boxes to supervise convolutional networks for semantic segmentation

    Jifeng Dai, Kaiming He, and Jian Sun. Boxsup: Exploit- ing bounding boxes to supervise convolutional networks for semantic segmentation. In Proceedings of the IEEE inter- national conference on computer vision, pages 1635–1643,

    work page

  6. [6]

    Mevis: A large-scale benchmark for video segmentation with motion expressions

    Henghui Ding, Chang Liu, Shuting He, Xudong Jiang, and Chen Change Loy. Mevis: A large-scale benchmark for video segmentation with motion expressions. In Proceedings of the IEEE/CVF international conference on computer vision, pages 2694–2703, 2023. 1

    work page 2023

  7. [7]

    Mose: A new dataset for video object segmentation in complex scenes

    Henghui Ding, Chang Liu, Shuting He, Xudong Jiang, Philip HS Torr, and Song Bai. Mose: A new dataset for video object segmentation in complex scenes. In Proceed- ings of the IEEE/CVF international conference on computer vision, pages 20224–20234, 2023. 1

    work page 2023

  8. [8]

    Semi-supervised semantic segmen- tation needs strong, varied perturbations

    Geoff French, Samuli Laine, Timo Aila, Michal Mackiewicz, and Graham Finlayson. Semi-supervised semantic segmen- tation needs strong, varied perturbations. arXiv preprint arXiv:1906.01916, 2019. 1

    arXiv 1906

Show all 55 references
  1. [9]

    Deep residual learning for image recognition

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In Proceed- ings of the IEEE conference on computer vision and pattern recognition, pages 770–778, 2016. 1, 6

    2016

  2. [10]

    Masked autoencoders are scalable vision learners

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll´ar, and Ross Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition , pages 16000– 16009, 2022. 1

    2022

  3. [11]

    Denoising diffu- sion probabilistic models

    Jonathan Ho, Ajay Jain, and Pieter Abbeel. Denoising diffu- sion probabilistic models. Advances in neural information processing systems, 33:6840–6851, 2020. 1, 3, 5

    2020

  4. [12]

    Fcns in the wild: Pixel-level adversarial and constraint-based adaptation

    Judy Hoffman, Dequan Wang, Fisher Yu, and Trevor Darrell. Fcns in the wild: Pixel-level adversarial and constraint-based adaptation. arXiv preprint arXiv:1612.02649, 2016. 1, 2

    2016 arXiv

  5. [13]

    Cycada: Cycle-consistent adversarial domain adaptation

    Judy Hoffman, Eric Tzeng, Taesung Park, Jun-Yan Zhu, Phillip Isola, Kate Saenko, Alexei Efros, and Trevor Dar- rell. Cycada: Cycle-consistent adversarial domain adaptation. In International conference on machine learning, pages 1989–

    1989

  6. [14]

    Three ways to improve semantic segmentation with self-supervised depth estimation

    Lukas Hoyer, Dengxin Dai, Yuhua Chen, Adrian Koring, Suman Saha, and Luc Van Gool. Three ways to improve semantic segmentation with self-supervised depth estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11130–11140, 2021. 1

    2021

  7. [15]

    Daformer: Improving network architectures and training strategies for domain-adaptive semantic segmentation

    Lukas Hoyer, Dengxin Dai, and Luc Van Gool. Daformer: Improving network architectures and training strategies for domain-adaptive semantic segmentation. arXiv preprint arXiv:2111.14887, 2021. 1, 2, 3, 7, 8

    2021 arXiv

  8. [16]

    Hrda: Context- aware high-resolution domain-adaptive semantic segmenta- tion

    Lukas Hoyer, Dengxin Dai, and Luc Van Gool. Hrda: Context- aware high-resolution domain-adaptive semantic segmenta- tion. arXiv preprint arXiv:2204.13132, 2022. 6, 7, 8

    2022 arXiv

  9. [17]

    Semi-supervised semantic seg- mentation with directional context-aware consistency

    Xin Lai, Zhuotao Tian, Li Jiang, Shu Liu, Hengshuang Zhao, Liwei Wang, and Jiaya Jia. Semi-supervised semantic seg- mentation with directional context-aware consistency. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1205–1214, 2021. 1

    2021

  10. [18]

    Handwritten digit recognition with a back-propagation net- work

    Yann LeCun, Bernhard Boser, John Denker, Donnie Hender- son, Richard Howard, Wayne Hubbard, and Lawrence Jackel. Handwritten digit recognition with a back-propagation net- work. Advances in neural information processing systems, 2,

  11. [19]

    Bidirectional learning for domain adaptation of semantic segmentation

    Yunsheng Li, Lu Yuan, and Nuno Vasconcelos. Bidirectional learning for domain adaptation of semantic segmentation. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 6936–6945, 2019. 1

    2019

  12. [20]

    Bapa-net: Boundary adaptation and prototype align- ment for cross-domain semantic segmentation

    Yahao Liu, Jinhong Deng, Xinchen Gao, Wen Li, and Lixin Duan. Bapa-net: Boundary adaptation and prototype align- ment for cross-domain semantic segmentation. In Proceed- ings of the IEEE/CVF International Conference on Computer Vision, pages 8801–8811, 2021. 6, 7, 8

    2021

  13. [21]

    Learn- ing deconvolution network for semantic segmentation

    Hyeonwoo Noh, Seunghoon Hong, and Bohyung Han. Learn- ing deconvolution network for semantic segmentation. In Proceedings of the IEEE international conference on com- puter vision, pages 1520–1528, 2015. 2

    2015

  14. [22]

    Classmix: Segmentation-based data aug- mentation for semi-supervised learning

    Viktor Olsson, Wilhelm Tranheden, Juliano Pinto, and Lennart Svensson. Classmix: Segmentation-based data aug- mentation for semi-supervised learning. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 1369–1378, 2021. 4, 5

    2021

  15. [23]

    Unsupervised domain adap- tation via domain-adaptive diffusion

    Duo Peng, Qiuhong Ke, ArulMurugan Ambikapathi, Yasin Yazici, Yinjie Lei, and Jun Liu. Unsupervised domain adap- tation via domain-adaptive diffusion. IEEE Transactions on Image Processing, 2024. 2

    2024

  16. [24]

    Learning target-domain-specific classifier for partial domain adaptation

    Chuan-Xian Ren, Pengfei Ge, Peiyi Yang, and Shuicheng Yan. Learning target-domain-specific classifier for partial domain adaptation. IEEE Transactions on Neural Networks and Learning Systems, 32(5):1989–2001, 2020. 1

    1989

  17. [25]

    Playing for data: Ground truth from computer games

    Stephan R Richter, Vibhav Vineet, Stefan Roth, and Vladlen Koltun. Playing for data: Ground truth from computer games. In European conference on computer vision, pages 102–118. Springer, 2016. 1, 2, 5, 6

    2016

  18. [26]

    High-resolution image synthesis with latent diffusion models

    Robin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser, and Bj ¨orn Ommer. High-resolution image synthesis with latent diffusion models. In Proceedings of 9 the IEEE/CVF conference on computer vision and pattern recognition, pages 10684–10695, 2022. 3, 6

    2022

  19. [27]

    U- net: Convolutional networks for biomedical image segmen- tation

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

    2015

  20. [28]

    The synthia dataset: A large collection of synthetic images for semantic segmentation of urban scenes

    German Ros, Laura Sellart, Joanna Materzynska, David Vazquez, and Antonio M Lopez. The synthia dataset: A large collection of synthetic images for semantic segmentation of urban scenes. In Proceedings of the IEEE conference on com- puter vision and pattern recognition, pages 3...

    2016

  21. [29]

    Adversarial learning approach for open set domain adaptation

    Kuniaki Saito, Shohei Yamamoto, Yoshitaka Ushiku, and Tatsuya Harada. Adversarial learning approach for open set domain adaptation. In Domain Adaptation in Computer Vision with Deep Learning, pages 175–193. Springer, 2020. 2

    2020

  22. [30]

    Fully convolutional networks for semantic segmentation

    Evan Shelhamer, Jonathan Long, and Trevor Darrell. Fully convolutional networks for semantic segmentation. IEEE transactions on pattern analysis and machine intelligence, 39 (4):640–651, 2016. 2

    2016

  23. [31]

    Box-driven class-wise region masking and filling rate guided loss for weakly supervised semantic segmentation

    Chunfeng Song, Yan Huang, Wanli Ouyang, and Liang Wang. Box-driven class-wise region masking and filling rate guided loss for weakly supervised semantic segmentation. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3136–3145, 2019. 1

    2019

  24. [32]

    Denoising diffusion implicit models

    Jiaming Song, Chenlin Meng, and Stefano Ermon. Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502,

    2010 arXiv

  25. [33]

    Score-based generative modeling through stochastic differential equations

    Yang Song, Jascha Sohl-Dickstein, Diederik P Kingma, Ab- hishek Kumar, Stefano Ermon, and Ben Poole. Score-based generative modeling through stochastic differential equations. arXiv preprint arXiv:2011.13456, 2020. 2

    2011 arXiv

  26. [34]

    Semi supervised semantic segmentation using generative adver- sarial network

    Nasim Souly, Concetto Spampinato, and Mubarak Shah. Semi supervised semantic segmentation using generative adver- sarial network. In Proceedings of the IEEE international conference on computer vision, pages 5688–5696, 2017. 1

    2017

  27. [35]

    Segmenter: Transformer for semantic segmentation

    Robin Strudel, Ricardo Garcia, Ivan Laptev, and Cordelia Schmid. Segmenter: Transformer for semantic segmentation. In Proceedings of the IEEE/CVF international conference on computer vision, pages 7262–7272, 2021. 2

    2021

  28. [36]

    Semantic diffusion network for semantic segmentation

    Haoru Tan, Sitong Wu, and Jimin Pi. Semantic diffusion network for semantic segmentation. Advances in Neural Information Processing Systems, 35:8702–8716, 2022. 2, 3

    2022

  29. [37]

    Un- supervised domain adaptation in semantic segmentation via orthogonal and clustered embeddings

    Marco Toldo, Umberto Michieli, and Pietro Zanuttigh. Un- supervised domain adaptation in semantic segmentation via orthogonal and clustered embeddings. In Proceedings of the IEEE/CVF Winter conference on Applications of Computer Vision, pages 1358–1368, 2021. 6

    2021

  30. [38]

    Dacs: Domain adaptation via cross- domain mixed sampling

    Wilhelm Tranheden, Viktor Olsson, Juliano Pinto, and Lennart Svensson. Dacs: Domain adaptation via cross- domain mixed sampling. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 1379–1389, 2021. 7

    2021

  31. [39]

    Learn- ing to adapt structured output space for semantic segmenta- tion

    Yi-Hsuan Tsai, Wei-Chih Hung, Samuel Schulter, Kihyuk Sohn, Ming-Hsuan Yang, and Manmohan Chandraker. Learn- ing to adapt structured output space for semantic segmenta- tion. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7472–7481, 201...

    2018

  32. [40]

    Cluda: Contrastive learning in unsupervised domain adaptation for semantic segmentation

    Midhun Vayyat, Jaswin Kasi, Anuraag Bhattacharya, Shuaib Ahmed, and Rahul Tallamraju. Cluda: Contrastive learning in unsupervised domain adaptation for semantic segmentation. arXiv preprint arXiv:2208.14227, 2022. 1, 6

    2022 arXiv

  33. [41]

    Reseg: A recurrent neural network-based model for semantic segmentation

    Francesco Visin, Marco Ciccone, Adriana Romero, Kyle Kast- ner, Kyunghyun Cho, Yoshua Bengio, Matteo Matteucci, and Aaron Courville. Reseg: A recurrent neural network-based model for semantic segmentation. In Proceedings of the IEEE conference on computer vision and pattern re...

    2016

  34. [42]

    Uctransnet: rethinking the skip connections in u-net from a channel-wise perspective with transformer

    Haonan Wang, Peng Cao, Jiaqi Wang, and Osmar R Zaiane. Uctransnet: rethinking the skip connections in u-net from a channel-wise perspective with transformer. In Proceedings of the AAAI conference on artificial intelligence, pages 2441– 2449, 2022. 4

    2022

  35. [43]

    Loveda: A remote sensing land-cover dataset for domain adaptive semantic segmentation

    Junjue Wang, Zhuo Zheng, Ailong Ma, Xiaoyan Lu, and Yanfei Zhong. Loveda: A remote sensing land-cover dataset for domain adaptive semantic segmentation. arXiv preprint arXiv:2110.08733, 2021. 1

    2021 arXiv

  36. [44]

    Domain adaptive semantic segmentation with self- supervised depth estimation

    Qin Wang, Dengxin Dai, Lukas Hoyer, Luc Van Gool, and Olga Fink. Domain adaptive semantic segmentation with self- supervised depth estimation. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 8515– 8525, 2021. 7

    2021

  37. [45]

    Cross-modality lge-cmr segmen- tation using image-to-image translation based data augmen- tation

    Wei Wang, Xinhua Yu, Bo Fang, Yue Zhao, Yongyong Chen, Wei Wei, and Junxin Chen. Cross-modality lge-cmr segmen- tation using image-to-image translation based data augmen- tation. IEEE/ACM Transactions on Computational Biology and Bioinformatics, 20(4):2367–2375, 2022. 2

    2022

  38. [46]

    Inet: convolutional networks for biomedical image segmentation

    Weihao Weng and Xin Zhu. Inet: convolutional networks for biomedical image segmentation. Ieee Access, 9:16591–16603,

  39. [47]

    Diffumask: Synthesizing images with pixel-level annotations for semantic segmentation using diffu- sion models

    Weijia Wu, Yuzhong Zhao, Mike Zheng Shou, Hong Zhou, and Chunhua Shen. Diffumask: Synthesizing images with pixel-level annotations for semantic segmentation using diffu- sion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 1206–1217, 2...

    2023

  40. [48]

    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. 2

    2021

  41. [49]

    Multi-source domain adaptation for unsupervised road defect segmentation

    Jongmin Yu, Hyeontaek Oh, Sebastiano Fichera, Paolo Pao- letti, and Shan Luo. Multi-source domain adaptation for unsupervised road defect segmentation. In 2023 IEEE In- ternational Conference on Robotics and Automation (ICRA), pages 5638–5644. IEEE, 2023. 1, 2

    2023

  42. [50]

    Adversar- ial denoising diffusion model for unsupervised anomaly de- tection

    Jongmin Yu, Hyeontaek Oh, and Jinhong Yang. Adversar- ial denoising diffusion model for unsupervised anomaly de- tection. In Deep Generative Models for Health Workshop NeurIPS 2023, 2023. 1, 2

    2023

  43. [51]

    Prototypical pseudo label denoising and target 10 structure learning for domain adaptive semantic segmentation

    Pan Zhang, Bo Zhang, Ting Zhang, Dong Chen, Yong Wang, and Fang Wen. Prototypical pseudo label denoising and target 10 structure learning for domain adaptive semantic segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 1241...

    2021

  44. [52]

    Curriculum domain adaptation for semantic segmentation of urban scenes

    Yang Zhang, Philip David, and Boqing Gong. Curriculum domain adaptation for semantic segmentation of urban scenes. In Proceedings of the IEEE international conference on com- puter vision, pages 2020–2030, 2017. 2

    2020

  45. [53]

    Unet++: Redesigning skip connections to exploit multiscale features in image segmen- tation

    Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jianming Liang. Unet++: Redesigning skip connections to exploit multiscale features in image segmen- tation. IEEE transactions on medical imaging, 39(6):1856– 1867, 2019. 4

    2019

  46. [54]

    Un- supervised domain adaptation for semantic segmentation via class-balanced self-training

    Yang Zou, Zhiding Yu, BVK Kumar, and Jinsong Wang. Un- supervised domain adaptation for semantic segmentation via class-balanced self-training. In Proceedings of the European conference on computer vision (ECCV), pages 289–305, 2018. 1, 7

    2018

  47. [55]

    Pseudoseg: Designing pseudo labels for semantic segmentation

    Yuliang Zou, Zizhao Zhang, Han Zhang, Chun-Liang Li, Xiao Bian, Jia-Bin Huang, and Tomas Pfister. Pseudoseg: Designing pseudo labels for semantic segmentation. arXiv preprint arXiv:2010.09713, 2020. 1 11

    2010 arXiv

Pith tools

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