pith. sign in

arxiv: 2605.24807 · v1 · pith:EDS5NTZCnew · submitted 2026-05-24 · 💻 cs.CV

CLIP-Guided SAM: Parameter-Efficient Semantic Conditioning for Promptable Segmentation

Shayan Jalilian , Abdul Bais This is my paper

Pith reviewed 2026-06-30 12:23 UTC · model grok-4.3

classification 💻 cs.CV
keywords CLIPSAMsemantic segmentationparameter-efficient fine-tuningvision-language modelspromptable segmentationmulti-modal adapterstext prompts
0
0 comments X

The pith

CLIP-Guided SAM injects text, vision, and similarity features from CLIP into SAM's image encoder via lightweight adapters to add semantic conditioning while keeping the original promptable interface intact.

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

SAM produces accurate masks from prompts but lacks built-in understanding of object categories. The method replaces external prompt generation with direct injection of CLIP features through small multi-modal adapters placed inside SAM's image encoder. This internal conditioning lets semantic signals shape mask predictions during both manual mixed-prompt use and text-only operation. The design targets low-labeled-data regimes and stresses that training and inference must use matching prompt types to maintain robustness. Experiments position the approach as competitive with parameter-efficient baselines and vision-language pipelines while adding few trainable weights.

Core claim

By injecting CLIP-derived text, vision, and similarity features directly into SAM's image encoder through lightweight multi-modal semantic adapters, the framework conditions internal feature representations so that semantic information influences mask prediction while preserving SAM's original promptable interface and supporting both manual and semi-automatic text-only modes.

What carries the argument

Lightweight multi-modal semantic adapters that inject CLIP-derived text, vision, and similarity features directly into SAM's image encoder to condition internal feature representations.

If this is right

  • Supports interactive segmentation that combines text and spatial prompts in manual mode.
  • Enables concept-specific segmentation from text input alone in semi-automatic mode.
  • Remains parameter-efficient during both training and inference across general and specialized tasks.
  • Requires consistent prompt types between training and inference to achieve reported robustness.
  • Delivers superior or competitive mask quality against SAM+PEFT baselines and vision-language pipelines while using limited labeled data.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The internal-conditioning pattern could be tested on other promptable foundation models that currently rely on external vision-language stages.
  • Prompt-type alignment during training may prove useful for any system that mixes textual and spatial inputs at inference time.
  • The low-labeled-data efficiency suggests direct applicability in domains where annotation is costly, such as specialized imaging tasks.
  • Replacing or augmenting the CLIP backbone with other vision-language encoders could be explored to vary the injected semantic signals.

Load-bearing premise

Directly injecting the CLIP features through the adapters will successfully improve semantic awareness without interfering with SAM's prompt-based mask generation or requiring extensive retraining.

What would settle it

On a general-domain benchmark such as COCO or a downstream task, CLIP-Guided SAM produces lower-quality masks or requires substantially more added parameters than a SAM model fine-tuned with standard PEFT methods that lack the semantic adapters.

Figures

Figures reproduced from arXiv: 2605.24807 by Abdul Bais, Shayan Jalilian.

Figure 1
Figure 1. Figure 1: CLIP-Guided SAM Overview. Comparison of (top) SAM with PEFT, (mid￾dle) typical VLM+SAM pipelines, and (bottom) our CLIP-Guided SAM. Existing ap￾proaches either rely on spatial prompting alone or couple a VLM to SAM externally via prompt generation, leaving SAM’s internal representations semantically uninformed. Our approach introduces internal semantic conditioning by injecting CLIP-derived features direct… view at source ↗
Figure 2
Figure 2. Figure 2: Semantic Prompt Generation with CLIP. Shows the process of how we use CLIP to generate our semantic prompts, which include similarity maps (dense mask prompt), vision+similarity features, text features, and point prompts. Point prompts are sampled either from similarity maps or GT labels, depending on the design mode (manual vs semi-automatic) yielding similarity scores s ∈ R N×1 . Reshaping s to H ×W give… view at source ↗
Figure 3
Figure 3. Figure 3: Overall integration framework. CLIP extracts text and image embeddings, which produce similarity features and maps via cosine similarity. Points prompts are derived either from similarity masks or GT labels, depending on our design mode (manual or semi-automatic), while text, vision, and similarity features are injected into SAM through our semantic adapter modules. and similarity scores s ∈ R N×1 . We fus… view at source ↗
Figure 4
Figure 4. Figure 4: Adapter Placement and Internal Structure. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Qualitative analysis on ADE20K (1/16 split). [PITH_FULL_IMAGE:figures/full_fig_p014_5.png] view at source ↗
read the original abstract

Promptable foundation models such as the Segment Anything Model (SAM) produce high-quality masks but remain semantically blind, relying on external prompts to specify categories. Existing vision-language approaches address this limitation by using external prompt coupling, where a vision-language model generates spatial prompts for SAM as a separate stage. We propose CLIP-Guided SAM, a parameter-efficient segmentation framework built on internal semantic conditioning. Instead of using semantic signals only to generate prompts, we inject CLIP-derived text, vision, and similarity features directly into SAM's image encoder through lightweight multi-modal semantic adapters. These adapters condition SAM's internal feature representations, allowing semantic information to influence mask prediction while preserving SAM's original promptable interface. Our framework is designed for low labeled-data settings and applies to both general-domain benchmarks and specialized downstream tasks. It supports two operating modes: Manual mode, for interactive segmentation with both text and spatial prompts, and Semi-Automatic text-only mode, for applications that require concept-specific segmentation using only textual input. We show that robustness depends on aligning training with the type of prompts used at inference, making train-test prompt consistency an important design principle. Through extensive experiments and ablations, we evaluate our method against SAM+PEFT baselines without semantic conditioning, vision-language + SAM pipelines, SAM 3, and strong semi-supervised segmentation methods that rely on large amounts of unlabeled data. Across these settings, CLIP-Guided SAM consistently achieves superior or competitive performance while remaining parameter-efficient in both training and deployment.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

0 major / 3 minor

Summary. The manuscript proposes CLIP-Guided SAM, a parameter-efficient framework that injects CLIP-derived text, vision, and similarity features directly into SAM's image encoder via lightweight multi-modal semantic adapters. This internal conditioning enables semantic awareness in promptable segmentation while preserving the original interface. The method supports manual (text+spatial) and semi-automatic (text-only) modes, emphasizes train-test prompt consistency as a design principle, and is evaluated on general benchmarks and specialized tasks against SAM+PEFT, VLM+SAM pipelines, SAM 3, and semi-supervised baselines, claiming superior or competitive results with efficiency in low-labeled-data regimes.

Significance. If the performance and efficiency claims hold under the reported conditions, the work provides a distinct internal-conditioning alternative to external prompt-coupling approaches, potentially improving semantic robustness in foundation models like SAM without full retraining. The explicit focus on prompt-consistency alignment and applicability to downstream tasks with limited labels adds practical value; the parameter-efficient design and dual-mode operation are clear strengths for deployment.

minor comments (3)
  1. [Abstract] Abstract: the phrase 'SAM 3' is ambiguous (likely a reference to a specific variant or baseline); clarify the exact model or citation in the main text and abstract.
  2. [§3] The description of adapter architecture and feature-injection points would benefit from an explicit equation or diagram reference early in §3 to make the multi-modal conditioning mechanism immediately reproducible from the text.
  3. [Experiments] Table captions or result sections should explicitly state the number of parameters updated during training for each compared method to strengthen the 'parameter-efficient' claim.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the positive summary, recognition of the method's strengths in internal semantic conditioning, prompt consistency, and efficiency, and the recommendation for minor revision. No major comments were listed in the report.

Circularity Check

0 steps flagged

No significant circularity; independent architectural proposal

full rationale

The provided abstract and description outline an independent architectural modification: lightweight multi-modal adapters that inject CLIP-derived features directly into SAM's image encoder to condition representations while preserving the original promptable interface. No equations, derivations, or self-referential definitions are supplied that would reduce any claimed result to fitted inputs or prior self-citations by construction. Performance claims rest on external experiments and ablations against baselines, with no indication that results are tautological or forced by the method's own definitions. The design is presented as a novel conditioning strategy rather than a renaming or self-citation-dependent uniqueness claim, making the derivation self-contained against the stated benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract provides no details on mathematical derivations, specific parameters, or background assumptions; no free parameters, axioms, or invented entities can be identified from available text.

pith-pipeline@v0.9.1-grok · 5805 in / 1007 out tokens · 50776 ms · 2026-06-30T12:23:25.164544+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

45 extracted references · 13 canonical work pages · 4 internal anchors

  1. [1]

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

    Aleem, S., Wang, F., Maniparambil, M., Arazo, E., Dietlmeier, J., Curran, K., Connor, N.E., Little, S.: Test-time adaptation with SALIP: A cascade of SAM and CLIP for zero-shot medical image segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 5184–5193 (2024)

    2024

  2. [2]

    SAM 3: Segment Anything with Concepts

    Carion, N., Gustafson, L., Hu, Y.T., Debnath, S., Hu, R., Suris, D., Ryali, C., Alwala, K.V., Khedr, H., Huang, A., Lei, J., Ma, T., Guo, B., Kalla, A., Marks, M., Greer, J., Wang, M., Sun, P., Rädle, R., Afouras, T., Mavroudi, E., Xu, K., Wu, T.H., Zhou, Y., Momeni, L., Hazra, R., Ding, S., Vaze, S., Porcher, F., Li, F., Li, S., Kamath, A., Cheng, H.K., ...

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  3. [3]

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

    Cha, J., Mun, J., Roh, B.: Learning to generate text-grounded mask for open- world semantic segmentation from only image-text pairs. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11165– 11174 (2023)

    2023

  4. [4]

    In: Proceedings of the IEEE/CVF International Conference on Computer Vision

    Chen, T., Zhu, L., Deng, C., Cao, R., Wang, Y., Zhang, S., Li, Z., Sun, L., Zang, Y., Mao, P.: Sam-adapter: Adapting segment anything in underperformed scenes. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 3367–3375 (2023)

    2023

  5. [5]

    arXiv preprint arXiv:2304.09148 (2023)

    Chen, T., Zhu, L., Ding, C., Cao, R., Wang, Y., Li, Z., Sun, L., Mao, P., Zang, Y.: Sam fails to segment anything?–sam-adapter: Adapting sam in underper- formed scenes: Camouflage, shadow, medical image segmentation, and more. arXiv preprint arXiv:2304.09148 (2023)

    work page arXiv 2023

  6. [6]

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

    Cho,S.,Shin,H.,Hong,S.,Arnab,A.,Seo,P.H.,Kim,S.:Cat-seg:Costaggregation for open-vocabulary semantic segmentation. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4113–4123 (2024)

    2024

  7. [7]

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

    Ding, J., Xue, N., Xia, G.S., Dai, D.: Decoupling zero-shot semantic segmenta- tion. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 11583–11592 (2022)

    2022

  8. [8]

    Fan, G.-P

    Fan, D.P., Ji, G.P., Cheng, M.M., Shao, L.: Concealed object detection. IEEE Transactions on Pattern Analysis and Machine Intelligence44(10), 6024–6042 (Oct 2022).https://doi.org/10.1109/tpami.2021.3085766,http://dx.doi.org/10. 1109/TPAMI.2021.3085766

    work page doi:10.1109/tpami.2021.3085766 2022

  9. [9]

    In: European conference on computer vision

    Ghiasi,G.,Gu,X.,Cui,Y.,Lin,T.Y.:Scalingopen-vocabularyimagesegmentation with image-level labels. In: European conference on computer vision. pp. 540–557. Springer (2022)

    2022

  10. [10]

    Open-vocabulary Object Detection via Vision and Language Knowledge Distillation

    Gu, X., Lin, T.Y., Kuo, W., Cui, Y.: Open-vocabulary object detection via vision and language knowledge distillation. arXiv preprint arXiv:2104.13921 (2021) 16 S. Jalilian and A. Bais

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  11. [11]

    In: International conference on machine learning

    Houlsby, N., Giurgiu, A., Jastrzebski, S., Morrone, B., De Laroussilhe, Q., Ges- mundo, A., Attariyan, M., Gelly, S.: Parameter-efficient transfer learning for nlp. In: International conference on machine learning. pp. 2790–2799. PMLR (2019)

    2019

  12. [12]

    In: European Conference on Computer Vision

    Hoyer, L., Tan, D.J., Naeem, M.F., Van Gool, L., Tombari, F.: Semivl: semi- supervised semantic segmentation with vision-language guidance. In: European Conference on Computer Vision. pp. 257–275. Springer (2024)

    2024

  13. [13]

    ICLR1(2), 3 (2022)

    Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., Chen, W., et al.: Lora: Low-rank adaptation of large language models. ICLR1(2), 3 (2022)

    2022

  14. [14]

    IEEE Access14, 31732–31746 (2026).https:// doi.org/10.1109/ACCESS.2026.3668182

    Jalilian, S., Bais, A.: Sam-ptx: Text-guided fine-tuning of sam with parameter- efficient, parallel-text adapters. IEEE Access14, 31732–31746 (2026).https:// doi.org/10.1109/ACCESS.2026.3668182

    work page doi:10.1109/access.2026.3668182 2026

  15. [15]

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

    Kirillov, A., Mintun, E., Ravi, N., Mao, H., Rolland, C., Gustafson, L., Xiao, T., Whitehead, S., Berg, A.C., Lo, W.Y., et al.: Segment anything. In: Proceedings of the IEEE/CVF international conference on computer vision. pp. 4015–4026 (2023)

    2023

  16. [16]

    arxiv 2024

    Koleilat, T., Asgariandehkordi, H., Rivaz, H., Xiao, Y.: Medclip-samv2: Towards universal text-driven medical image segmentation. arxiv 2024. arXiv preprint arXiv:2409.19483 (2024)

    work page arXiv 2024

  17. [17]

    arXiv preprint arXiv:2407.15728 (2024)

    Kollias, D., Arsenos, A., Wingate, J., Kollias, S.: Sam2clip2sam: Vision language model for segmentation of 3d ct scans for covid-19 detection. arXiv preprint arXiv:2407.15728 (2024)

    work page arXiv 2024

  18. [18]

    Computer vision and image understanding 184, 45–56 (2019)

    Le, T.N., Nguyen, T.V., Nie, Z., Tran, M.T., Sugimoto, A.: Anabranch network for camouflaged object segmentation. Computer vision and image understanding 184, 45–56 (2019)

    2019

  19. [19]

    The Power of Scale for Parameter-Efficient Prompt Tuning

    Lester, B., Al-Rfou, R., Constant, N.: The power of scale for parameter-efficient prompt tuning. arXiv preprint arXiv:2104.08691 (2021)

    work page internal anchor Pith review Pith/arXiv arXiv 2021

  20. [20]

    Language-driven Semantic Segmentation

    Li, B., Weinberger, K.Q., Belongie, S., Koltun, V., Ranftl, R.: Language-driven semantic segmentation. arXiv preprint arXiv:2201.03546 (2022)

    work page internal anchor Pith review Pith/arXiv arXiv 2022

  21. [21]

    Neurocomputing618, 129122 (2025)

    Li, S., Cao, J., Ye, P., Ding, Y., Tu, C., Chen, T.: ClipSAM: CLIP and SAM collaboration for zero-shot anomaly segmentation. Neurocomputing618, 129122 (2025)

    2025

  22. [22]

    Pattern Recognition162, 111409 (2025)

    Li, Y., Wang, H., Duan, Y., Zhang, J., Li, X.: A closer look at the explainability of contrastive language-image pre-training. Pattern Recognition162, 111409 (2025)

    2025

  23. [23]

    In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition

    Liang, F., Wu, B., Dai, X., Li, K., Zhao, Y., Zhang, H., Zhang, P., Vajda, P., Mar- culescu, D.: Open-vocabulary semantic segmentation with mask-adapted clip. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recog- nition. pp. 7061–7070 (2023)

    2023

  24. [24]

    arXiv preprint arXiv:2503.22237 (2025)

    Liu, K., Wang, J., Jin, R., Hwang, W., Chung, T.S.: Schnet: Sam marries clip for human parsing. arXiv preprint arXiv:2503.22237 (2025)

    work page arXiv 2025

  25. [25]

    In: European conference on computer vision

    Liu, S., Zeng, Z., Ren, T., Li, F., Zhang, H., Yang, J., Jiang, Q., Li, C., Yang, J., Su, H., et al.: Grounding dino: Marrying dino with grounded pre-training for open-set object detection. In: European conference on computer vision. pp. 38–55. Springer (2024)

    2024

  26. [26]

    In: International Conference on Machine Learning

    Luo, H., Bao, J., Wu, Y., He, X., Li, T.: Segclip: Patch aggregation with learnable centers for open-vocabulary semantic segmentation. In: International Conference on Machine Learning. pp. 23033–23044. PMLR (2023)

    2023

  27. [27]

    In: 2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI)

    Ma, X., Fu, J., Liao, W., Zhang, S., Wang, G.: Clisc: Bridging clip and sam by enhanced cam for unsupervised brain tumor segmentation. In: 2025 IEEE 22nd International Symposium on Biomedical Imaging (ISBI). pp. 1–5. IEEE (2025)

    2025

  28. [28]

    In: International conference on machine learning

    Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., Clark, J., et al.: Learning transferable visual models from CLIP-Guided SAM 17 natural language supervision. In: International conference on machine learning. pp. 8748–8763. PmLR (2021)

    2021

  29. [29]

    Ren, T., Liu, S., Zeng, A., Lin, J., Li, K., Cao, H., Chen, J., Huang, X., Chen, Y., Yan, F., Zeng, Z., Zhang, H., Li, F., Yang, J., Li, H., Jiang, Q., Zhang, L.: Grounded sam: Assembling open-world models for diverse visual tasks (2024)

    2024

  30. [30]

    Advances in Neural Information Processing Systems35, 33754–33767 (2022)

    Shin,G.,Xie,W.,Albanie,S.:Reco:Retrieveandco-segmentforzero-shottransfer. Advances in Neural Information Processing Systems35, 33754–33767 (2022)

    2022

  31. [31]

    Skurowski, P., Abdulameer, H., Błaszczyk, J., Depta, T., Kornacki, A., Kozieł, P.: Animal camouflage analysis: Chameleon database2(6), 7

  32. [32]

    arXiv preprint arXiv:2401.17803 (2024)

    Song, Y., Zhou, Q., Lu, X., Shao, Z., Ma, L.: Su-sam: A simple unified framework for adapting segment anything model in underperformed scenes. arXiv preprint arXiv:2401.17803 (2024)

    work page arXiv 2024

  33. [33]

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

    Wang, H., Vasu, P.K.A., Faghri, F., Vemulapalli, R., Farajtabar, M., Mehta, S., Rastegari, M., Tuzel, O., Pouransari, H.: Sam-clip: Merging vision foundation mod- els towards semantic and spatial understanding. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 3635–3647 (2024)

    2024

  34. [34]

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

    Xu, J., De Mello, S., Liu, S., Byeon, W., Breuel, T., Kautz, J., Wang, X.: Groupvit: Semantic segmentation emerges from text supervision. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 18134– 18144 (2022)

    2022

  35. [35]

    In: European Conference on Computer Vision

    Xu, M., Zhang, Z., Wei, F., Lin, Y., Cao, Y., Hu, H., Bai, X.: A simple baseline for open-vocabulary semantic segmentation with pre-trained vision-language model. In: European Conference on Computer Vision. pp. 736–753. Springer (2022)

    2022

  36. [36]

    Yang,L.,Qi,L.,Feng,L.,Zhang,W.,Shi,Y.:Revisitingweak-to-strongconsistency in semi-supervised semantic segmentation (2023),https://arxiv.org/abs/2208. 09910

    2023

  37. [37]

    IEEE Transactions on Pattern Analysis and Machine In- telligence (2025)

    Yang, L., Zhao, Z., Zhao, H.: Unimatch v2: Pushing the limit of semi-supervised semantic segmentation. IEEE Transactions on Pattern Analysis and Machine In- telligence (2025)

    2025

  38. [38]

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

    Yang, X., Gong, X.: Foundation model assisted weakly supervised semantic seg- mentation. In: Proceedings of the IEEE/CVF winter conference on applications of computer vision. pp. 523–532 (2024)

    2024

  39. [39]

    In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

    Yu, X., Elazab, A., Ge, R., Jin, H., Jiang, X., Jia, G., Wu, Q., Shi, Q., Wang, C.: Ich-scnet: Intracerebral hemorrhage segmentation and prognosis classification network using clip-guided sam mechanism. In: 2024 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). pp. 2795–2800. IEEE (2024)

    2024

  40. [40]

    In: 2024 IEEE International Conference on Multimedia and Expo (ICME)

    Yu, Y., Xu, C., Wang, K.: Ts-sam: Fine-tuning segment-anything model for down- stream tasks. In: 2024 IEEE International Conference on Multimedia and Expo (ICME). pp. 1–6. IEEE (2024)

    2024

  41. [41]

    In: European Conference on Computer Vision

    Yuan, H., Li, X., Zhou, C., Li, Y., Chen, K., Loy, C.C.: Open-vocabulary sam: Seg- ment and recognize twenty-thousand classes interactively. In: European Conference on Computer Vision. pp. 419–437. Springer (2024)

    2024

  42. [42]

    Zhong, Z

    Zhong, Z., Tang, Z., He, T., Fang, H., Yuan, C.: Convolution meets lora: Parameter efficient finetuning for segment anything model. arXiv preprint arXiv:2401.17868 (2024)

    work page arXiv 2024

  43. [43]

    In: European conference on computer vision

    Zhou, C., Loy, C.C., Dai, B.: Extract free dense labels from clip. In: European conference on computer vision. pp. 696–712. Springer (2022)

    2022

  44. [44]

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

    Zhou, Z., Lei, Y., Zhang, B., Liu, L., Liu, Y.: Zegclip: Towards adapting clip for zero-shot semantic segmentation. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. pp. 11175–11185 (2023) 18 S. Jalilian and A. Bais

    2023

  45. [45]

    arXiv preprint arXiv:2010.09713 (2020)

    Zou, Y., Zhang, Z., Zhang, H., Li, C.L., Bian, X., Huang, J.B., Pfister, T.: Pseudoseg: Designing pseudo labels for semantic segmentation. arXiv preprint arXiv:2010.09713 (2020)

    work page arXiv 2010