Pith. sign in

REVIEW 4 major objections 6 minor 45 references

Exploring Palette based Color Guidance in Diffusion Models

T0 review · 4 major / 6 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read This paper claims that encoding color palettes as learned embeddings and fusing them with text in cross-attention gives diffusion models global color-scheme control that text prompts alone cannot provide.

desk verdict A solid but incremental empirical paper on learned palette embeddings for diffusion color control, with two evaluation blind spots: oracle-conditioned image-palette results and no user-defined palettes tested. read the letter →

arxiv 2508.08754 v1 pith:SQR2KFJM submitted 2025-08-12 cs.GR cs.CVcs.MM

Qianru Qiu , Jiafeng Mao , Xueting Wang This is my paper · ORCID
classification cs.GRcs.CVcs.MM
keywords colorguidancediffusionmodelspaletterepresentationimagecolorizationcross-attentionconditioningtext-to-imagegenerationuserstudy
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 tries to establish that a color palette should be a conditioning input in its own right for diffusion-based image generation, rather than being described in words or appended to the prompt. It proposes a palette-guided colorization framework in which a learned palette embedding is fused with the text embedding and injected through cross-attention, while the grayscale image is supplied through a separate structural branch. If the claim holds, users gain global control over the color scheme of a generated image, including backgrounds and objects never named in the prompt, by choosing a palette or pointing to a reference image. The paper reports that this approach outperforms text-only and palette-in-prompt baselines on color-alignment metrics and in a preference study with professional designers.

What carries the argument

The load-bearing component is the learned palette embedding. A palette is tokenized as discrete color codes in a $16\times16\times16$ CIELAB histogram, embedded, and processed by a masked color model that predicts masked color tokens; optional text or image conditions are mixed in through cross-attention. The resulting palette embedding is projected into the same space as the text embedding, concatenated with it, and fed into the cross-attention layers of the diffusion U-Net. This makes the palette a first-class conditioning signal, so the model can apply the palette globally rather than only to objects named in the prompt.

What would settle it

Take hand-authored palettes with deliberately non-natural color combinations, such as a brand palette or a neon-only palette, run the text-palette model and the palette-in-prompt baselines on the same grayscale images, and compare DCCW and histogram distance on the outputs; if the learned-embedding advantage disappears or reverses on these palettes, the claim of general palette-guided color control is falsified.

Watch

Extended reading notes

Core claim

The central discovery is that raw palette colors only help a diffusion model when they are converted into a learned embedding; writing RGB values into the prompt is not just weak, it can degrade output. The paper demonstrates this in an image-colorization setting: a grayscale image enters through a spatial-control branch, while a concatenation of text and palette embeddings is injected through cross-attention in the U-Net. Palette embeddings are produced by a masked color model trained over color tokens in CIELAB space, optionally conditioned on text or on a reference image. Across the LAION-derived and COCO-Stuff-derived test sets, palette-guided models improve histogram distance and palett

Load-bearing premise

The method is trained and evaluated only on five-color palettes extracted by K-means from natural images, so the promised benefit for user-customized palettes depends on the learned embeddings generalizing to palettes unlike those found in natural photos.

Editorial extensions

If this is right

  • Users could control the color mood of an entire image, backgrounds and minor objects included, by selecting a palette without having to name every colored element in the prompt.
  • Palette information must be embedded, not spelled out: adding RGB values to prompts can make color alignment worse than using no palette at all.
  • Palettes extracted from reference images transfer color style while leaving texture alone, which style-transfer methods that entangle color and texture do not offer.
  • Conditioning palette representations on text or images, especially with cross-modal encoders, yields sharper color predictions than palette-only representations.
  • The same palette-text fusion can be shifted to other color-sensitive generation tasks, such as text-to-image generation, where global color control is also missing.

Reading between the lines

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

  • A direct next step is to make text-versus-palette priority user-controllable; the paper documents cases where text wins over palette but leaves the arbitration mechanism unspecified.
  • Because the palette and text embeddings are concatenated before cross-attention, the fusion should transfer to text-to-image generation without the grayscale branch; that is a testable extension the paper motivates but does not run.
  • Training the palette representation on hand-authored palettes rather than only K-means-extracted natural-image palettes would test whether the method serves professional design palettes; the current datasets leave this open.
  • The same masked color model could double as an automatic color-harmony checker, since it already predicts plausible palette colors from partial context.
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

4 major / 6 minor

Summary. The paper proposes a palette-based guidance mechanism for diffusion-based image colorization. The authors construct two palette-text-image datasets (LAION-22k-PTI and COCO-Stuff-PTI) and train three palette representation models — palette-only, text-palette (with CLIP or SBERT text encoders), and image-palette (with CLIP or DINOv2 image encoders) — based on an extended masked color model. These representations are fused with text embeddings and injected into a Stable Diffusion-based colorization model via cross-attention, while the grayscale input is fed through a ControlNet branch. The main claim (Section 5.2, Tables 3 and 4) is that learned palette embeddings significantly improve color control over text-only and palette-in-prompt baselines on color-level metrics (Hist_Bha, DCCW), with a user study involving 33 designers supporting the qualitative preference for the proposed method.

Significance. If the results hold, this is a useful contribution to controllable colorization: it introduces a new conditioning modality (palette) and systematically compares several palette representation strategies, including a user study with professional designers. The datasets and the exploration of text/image encoders for palette representation are valuable assets. The paper is clearly motivated and the experimental design is mostly appropriate, but the evaluation has a few load-bearing gaps (oracle reference for the image-palette model, no hand-authored palettes, no statistical significance tests, and no simple learned-palette baseline). These issues prevent the paper from fully supporting its central claims in its current form, but they are addressable with additional experiments.

major comments (4)
  1. [Section 4.4 / Figure 2 caption] The image-palette model is evaluated with the ground-truth color image as the reference image. This follows from the dataset construction (palettes are K-means extracted from the original image) and from the reported PSNR/SSIM/LPIPS against ground truth. Consequently, the large gains of the image-palette variant in Tables 3 and 4 (e.g., DCCW 14.12 vs. 16.07 on LAION) conflate the effect of the learned palette embedding with the oracle reference-image condition. The claim that the image-palette representation is superior is uninterpretable unless a non-oracle reference image is used at inference, or an ablation removes the reference-image branch while keeping the palette.
  2. [Section 4.1 / Section 5.2] All evaluation palettes are K-means clusters extracted from natural images, and the claimed 'user-selected palette' scenario (Figure 1, Section 3.1) is never tested. The model is trained and evaluated on the same distribution of natural-image K-means palettes, so the paper does not support the practical claim that hand-authored palettes with out-of-distribution colors (e.g., strong neon or pastel palettes) will be faithfully followed. I recommend adding an evaluation with user-designed palettes or at least synthetic palettes sampled from a broader CIELAB distribution, and reporting whether the colorization outputs align with those palettes on Hist_Bha and DCCW.
  3. [Tables 3 and 4] No error bars, confidence intervals, or significance tests are reported for any colorization metric. Since the generation process is stochastic (DDIM sampling) and several differences are small (e.g., LPIPS 0.3601 vs. 0.3671 on LAION; SSIM consistently lower than L-CAD), the abstract's claim of 'significantly improves' is not statistically supported. Please report means and standard deviations over multiple seeds or inference runs, and perform significance tests (e.g., paired bootstrap or Wilcoxon) for the key color-level metrics.
  4. [Section 4.3 / Tables 3 and 4] The only palette-conditioned baselines are palette descriptions inserted into the text prompt, which is a deliberately weak baseline given CLIP's poor handling of raw RGB values. To isolate the contribution of the masked-color representation, the paper should include a simple learned baseline, such as a linear or MLP projection of the five palette RGB values into the same cross-attention conditioning space. Without this, it is unclear whether the observed improvement comes from the proposed palette representation or merely from using any learned embedding of palette values.
minor comments (6)
  1. [Figure 3 caption] Typo: 'boarder' should be 'broader'.
  2. [Section 3.2] The concatenation of palette and text embeddings is described only as 'with aligned dimensions' and 'projected into the same space.' Please specify whether the concatenation is along the sequence dimension (adding tokens) or the feature dimension (pooled vector), and give the resulting conditioning vector shape in the cross-attention layers.
  3. [Section 4.2] Hyperparameters for the palette representation models (learning rate, batch size, number of epochs, sequence length, token vocabulary size) are not reported. At least the sequence length and vocabulary size should be stated for reproducibility.
  4. [Section 5.3] The text says 33 designers were recruited, but Table 5 reports 28 responses. Please clarify whether 5 did not complete the questionnaire or were excluded.
  5. [Section 3.3] The phrase 'textural data' should likely be 'textual data' when referring to Sentence-BERT training.
  6. [Section 4.4] The DCCW metric is used both for palette-model evaluation and for colorization evaluation. It would help to state explicitly whether DCCW is computed between the predicted/generated palette and the reference palette after dynamic warping, and whether the generated palette is also K-means extracted with the same K=5.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: the palette-guidance framework is trained and evaluated empirically; the self-citations are transparent and not load-bearing. The main concerns are evaluation-validity gaps, not derivation circularity.

full rationale

The paper makes no first-principles derivation that reduces a predicted quantity to a fitted parameter. The palette representation models are trained end-to-end on a masked-color objective and separately evaluated on held-out test sets (Tables 1–2); this is standard supervised evaluation, not circular prediction. The colorization models are also trained and tested on held-out images with quantitative color metrics (Tables 3–4), so the central claim that learned palette embeddings improve color control has independent empirical content. The only notable self-citations are [22,23], the authors' prior masked-color modeling work, cited transparently in Sections 2.1 and 3.3 ('we extend the multimodal masked color modeling approach proposed by Qiu et al. [22]'). This is a building-block citation, not an imported uniqueness theorem or an ansatz smuggled in to forbid alternatives. The paper's real weaknesses are external-validity gaps, which I flag rather than treating as circularity: (1) Section 4.1 says 'From each image, we extracted a five-color palette using K-means clustering', meaning all quantitative tests use natural-image palettes; the claimed user-customized-palette benefit (Section 3.1, Figure 1) is never quantitatively tested with hand-authored palettes. (2) Figure 2's caption says 'During the training process, the reference image is the original input image, while a different image can be used in the inference process', but the paper does not report whether Tables 3–4 use the original image as the reference for the image-palette row; if so, the strong PSNR/SSIM gains could reflect reference-image leakage. (3) Section 5.4 honestly lists 'Color bleeding' and 'Conflict between text and palette guidance' as limitations, but does not mention the absence of user-palette evaluation. These are evaluation-design concerns, not cases where the claimed result is equivalent to its input by construction. Therefore the circularity score is low.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The paper is fully empirical: every component is either a pretrained model, a learned module, or a dataset construction choice. The central claim rests on the ability of learned palette embeddings to convey color information, which is validated only within the same K-means palette-extraction pipeline used for training. No new physical or ontological entities are introduced.

free parameters (6)
  • palette_size_K = 5
    Number of colors extracted per image palette; used for all datasets and models. The central claim may depend on this resolution.
  • CIELAB_bin_count_b = 16
    Color quantization resolution (b x b x b bins) for palette tokens, propagated from Qiu et al. [22]. Coarser or finer binning changes palette embedding capacity.
  • embedding_dim = 768
    Color token and projection dimension, matched to CLIP embeddings for concatenation.
  • checkpoint_selection = epochs 15, 20, 30
    Colorization checkpoints selected by validation performance; the reported scores depend on this selection.
  • DDIM_steps = 50
    Sampling steps used for evaluation; affects generation quality but is a standard choice.
  • learning_rate = 1e-5
    Learning rate for colorization training; typical but chosen by hand.
assumptions (5)
  • domain assumption Stable Diffusion 1.5 provides a valid pretrained generative backbone
    The colorization model is built on SD 1.5; the method inherits its biases and capabilities (Section 4.2).
  • domain assumption Masked color modeling on quantized CIELAB tokens yields useful palette representations
    Adopted from Qiu et al. [22]; the paper does not analyze whether the learned embeddings are sufficient for downstream color guidance.
  • domain assumption CLIP text and image encoders capture color semantics relevant to palette representation
    The paper selects CLIP over SBERT/DINOv2 based on masked-color accuracy, assuming this translates to colorization quality (Section 5.1).
  • domain assumption K-means in CIELAB with K=5 extracts the primary colors of an image
    Used to build all palettes in datasets and evaluation; the method is only tested with palettes produced by this procedure (Section 4.1).
  • domain assumption ControlNet-style branch can convey grayscale spatial structure without interfering with color conditioning
    The framework uses a ControlNet branch to inject structure; the paper relies on its ability to isolate structure from color (Section 3.2).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Exploring Palette based Color Guidance in Diffusion Models." pith.science (2026). https://pith.science/paper/SQR2KFJM

@misc{pith2026250808754,
  author       = {Pith},
  title        = {Pith review of: Exploring Palette based Color Guidance in Diffusion Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/SQR2KFJM}},
  note         = {Machine review of arXiv:2508.08754}
}
read the original abstract

With the advent of diffusion models, Text-to-Image (T2I) generation has seen substantial advancements. Current T2I models allow users to specify object colors using linguistic color names, and some methods aim to personalize color-object association through prompt learning. However, existing models struggle to provide comprehensive control over the color schemes of an entire image, especially for background elements and less prominent objects not explicitly mentioned in prompts. This paper proposes a novel approach to enhance color scheme control by integrating color palettes as a separate guidance mechanism alongside prompt instructions. We investigate the effectiveness of palette guidance by exploring various palette representation methods within a diffusion-based image colorization framework. To facilitate this exploration, we construct specialized palette-text-image datasets and conduct extensive quantitative and qualitative analyses. Our results demonstrate that incorporating palette guidance significantly improves the model's ability to generate images with desired color schemes, enabling a more controlled and refined colorization process.

Figures

Figures reproduced from arXiv: 2508.08754 by the authors.

Figure 1
Figure 1. Colorized results produced by diffusion models under palette guidance conditions. The grayscale image and text [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The framework for exploring various palette representation guidance in a diffusion-based image colorization model. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Color distribution of the LAION-22k-PTI and COCO [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Comparison of colorization results using different methods. ControlNet and L-CAD refer to the original models using [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: Colorization results of our method using palette [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Evaluation interface with reference palette. Op [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 8
Figure 8. Figure 8: Evaluation results with reference palette. [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 32 canonical work pages

  1. [22]

    Qianru Qiu, Xueting Wang, and Mayu Otani. 2023. Multimodal Color Recommen- dation in Vector Graphic Documents. InProceedings of the 31st ACM International Conference on Multimedia. 4003–4011

    work page 2023

  2. [1]

    Frank J Aherne, Neil A Thacker, and Peter I Rockett. 1998. The Bhattacharyya metric as an absolute similarity measure for frequency coded data. Kybernetika 34, 4 (1998), 363–368

    work page 1998

  3. [2]

    Yuval Alaluf, Daniel Garibi, Or Patashnik, Hadar Averbuch-Elor, and Daniel Cohen-Or. 2024. Cross-image attention for zero-shot appearance transfer. In ACM SIGGRAPH 2024 Conference Papers . 1–12

    work page 2024

  4. [3]

    Muhammad Atif Butt, Kai Wang, Javier Vazquez-Corral, and Joost van de Weijer

    work page

  5. [4]

    Piotr Bojanowski, Edouard Grave, Armand Joulin, and Tomas Mikolov. 2017. Enriching word vectors with subword information.Transactions of the association for computational linguistics 5 (2017), 135–146. Exploring Palette based Color Guidance in Diffusion Models *, *,

    work page 2017

  6. [5]

    Holger Caesar, Jasper Uijlings, and Vittorio Ferrari. 2018. Coco-stuff: Thing and stuff classes in context. In Proceedings of the IEEE conference on computer vision and pattern recognition. 1209–1218

    work page 2018

  7. [6]

    Huiwen Chang, Ohad Fried, Yiming Liu, Stephen DiVerdi, and Adam Finkelstein

    work page

  8. [7]

    Zheng Chang, Shuchen Weng, Yu Li, Si Li, and Boxin Shi. 2022. L-CoDer: Language-based colorization with color-object decoupling transformer. In Euro- pean Conference on Computer Vision . Springer, 360–375

    work page 2022

Show all 45 references
  1. [8]

    Zheng Chang, Shuchen Weng, Peixuan Zhang, Yu Li, Si Li, and Boxin Shi. 2023. L-CoIns: Language-based colorization with instance awareness. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 19221– 19230

    2023

  2. [9]

    Junho Cho, Sangdoo Yun, Kyoung Mu Lee, and Jin Young Choi. 2017. Palettenet: Image recolorization with given color palette. InProceedings of the ieee conference on computer vision and pattern recognition workshops . 62–70

    2017

  3. [10]

    Jiwoo Chung, Sangeek Hyun, and Jae-Pil Heo. 2024. Style injection in diffusion: A training-free approach for adapting large-scale diffusion models for style transfer. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 8795–8805

    2024

  4. [11]

    Zhengjun Du, Kai-Xiang Lei, Kun Xu, Jianchao Tan, and Yotam I Gingold. 2021. Video recoloring via spatial-temporal geometric palettes. ACM Trans. Graph. 40, 4 (2021), 150–1

    2021

  5. [12]

    Bingchen Gong, Yuehao Wang, Xiaoguang Han, and Qi Dou. 2023. RecolorNeRF: Layer decomposed radiance fields for efficient color editing of 3D scenes. In Proceedings of the 31st ACM International Conference on Multimedia . 8004–8015

    2023

  6. [13]

    Quan Huynh-Thu and Mohammed Ghanbari. 2008. Scope of validity of PSNR in image/video quality assessment. Electronics letters 44, 13 (2008), 800–801

    2008

  7. [14]

    Eunseo Kim, Jeongmin Hong, Hyuna Lee, and Minsam Ko. 2022. Colorbo: Envi- sioned mandala coloringthrough human-ai collaboration. In Proceedings of the 27th International Conference on Intelligent User Interfaces . 15–26

    2022

  8. [15]

    Suzi Kim and Sunghee Choi. 2021. Dynamic closest color warping to sort and compare palettes. ACM Transactions on Graphics (TOG) 40, 4 (2021), 1–15

    2021

  9. [16]

    Alexander Kirillov, Eric Mintun, Nikhila Ravi, Hanzi Mao, Chloe Rolland, Laura Gustafson, Tete Xiao, Spencer Whitehead, Alexander C Berg, Wan-Yen Lo, et al

  10. [17]

    Naoki Kita and Kazunori Miyata. 2016. Aesthetic rating and color suggestion for color palettes. In Computer Graphics Forum , Vol. 35. Wiley Online Library, 127–136

    2016

  11. [18]

    Zhengfei Kuang, Fujun Luan, Sai Bi, Zhixin Shu, Gordon Wetzstein, and Kalyan Sunkavalli. 2023. Palettenerf: Palette-based appearance editing of neural radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 20691–20700

    2023

  12. [19]

    Peter O’Donovan, Aseem Agarwala, and Aaron Hertzmann. 2011. Color compat- ibility from large datasets. In ACM SIGGRAPH 2011 papers. 1–12

    2011

  13. [20]

    Maxime Oquab, Timothée Darcet, Théo Moutakanni, Huy Vo, Marc Szafraniec, Vasil Khalidov, Pierre Fernandez, Daniel Haziza, Francisco Massa, Alaaeldin El- Nouby, et al. 2023. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193 (2023)

    2023 arXiv

  14. [21]

    Dustin Podell, Zion English, Kyle Lacey, Andreas Blattmann, Tim Dockhorn, Jonas Müller, Joe Penna, and Robin Rombach. 2023. Sdxl: Improving latent diffusion models for high-resolution image synthesis. arXiv preprint arXiv:2307.01952 (2023)

    2023 arXiv

  15. [23]

    Qianru Qiu, Xueting Wang, Mayu Otani, and Yuki Iwazaki. 2023. Color recom- mendation for vector graphic documents based on multi-palette representation. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 3621–3629

    2023

  16. [24]

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. 2021. Learning transferable visual models from natural language supervision. In International conference on machine learni...

    2021

  17. [25]

    Nils Reimers and Iryna Gurevych. 2019. Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. In Conference on Empirical Methods in Natural Language Processing. https://api.semanticscholar.org/CorpusID:201646309

    2019

  18. [26]

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

    2022

  19. [27]

    Christoph Schuhmann, Romain Beaumont, Richard Vencu, Cade Gordon, Ross Wightman, Mehdi Cherti, Theo Coombes, Aarush Katta, Clayton Mullis, Mitchell Wortsman, Patrick Schramowski, Srivatsa Kundurthy, Katherine Crowson, Ludwig Schmidt, Robert Kaczmarczyk, and Jenia Jitsev. 2022....

    2022 arXiv

  20. [28]

    Christoph Schuhmann, Richard Vencu, Romain Beaumont, Robert Kaczmarczyk, Clayton Mullis, Aarush Katta, Theo Coombes, Jenia Jitsev, and Aran Komatsuzaki

  21. [29]

    Jianchao Tan, Jose Echevarria, and Yotam Gingold. 2018. Efficient palette-based decomposition and recoloring of images via RGBXY-space geometry. ACM Transactions on Graphics (TOG) 37, 6 (2018), 1–10

    2018

  22. [30]

    Zhou Wang, Alan C Bovik, Hamid R Sheikh, and Eero P Simoncelli. 2004. Image quality assessment: from error visibility to structural similarity.IEEE transactions on image processing 13, 4 (2004), 600–612

    2004

  23. [31]

    Zhizhong Wang, Lei Zhao, and Wei Xing. 2023. Stylediffusion: Controllable disentangled style transfer via diffusion models. In Proceedings of the IEEE/CVF International Conference on Computer Vision . 7677–7689

    2023

  24. [32]

    Shuchen Weng, Hao Wu, Zheng Chang, Jiajun Tang, Si Li, and Boxin Shi. 2022. L-CoDe: Language-based colorization using color-object decoupled conditions. In Proceedings of the AAAI Conference on Artificial Intelligence , Vol. 36. 2677–2684

    2022

  25. [33]

    Shuchen Weng, Peixuan Zhang, Yu Li, Si Li, Boxin Shi, et al . 2024. L-CAD: Language-based Colorization with Any-level Descriptions using Diffusion Priors. Advances in Neural Information Processing Systems 36 (2024)

    2024

  26. [34]

    Yi Xiao, Peiyao Zhou, Yan Zheng, and Chi-Sing Leung. 2019. Interactive deep colorization using simultaneous global and local inputs. In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) . IEEE, 1887–1891

    2019

  27. [35]

    Zhongyou Xu, Tingting Wang, Faming Fang, Yun Sheng, and Guixu Zhang. 2020. Stylization-based architecture for fast deep exemplar colorization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition . 9363– 9372

    2020

  28. [36]

    Attention Is All You Need

    Wang Yin, Peng Lu, Zhaoran Zhao, and Xujun Peng. 2021. Yes, " Attention Is All You Need", for Exemplar based Colorization. In Proceedings of the 29th ACM international conference on multimedia . 2243–2251

    2021

  29. [37]

    Nir Zabari, Aharon Azulay, Alexey Gorkor, Tavi Halperin, and Ohad Fried. 2023. Diffusing colors: Image colorization with text guided diffusion. In SIGGRAPH Asia 2023 Conference Papers . 1–11

    2023

  30. [38]

    Lvmin Zhang, Anyi Rao, and Maneesh Agrawala. 2023. Adding conditional con- trol to text-to-image diffusion models. InProceedings of the IEEE/CVF International Conference on Computer Vision . 3836–3847

    2023

  31. [39]

    Richard Zhang, Phillip Isola, Alexei A Efros, Eli Shechtman, and Oliver Wang

  32. [40]

    Yuxin Zhang, Nisha Huang, Fan Tang, Haibin Huang, Chongyang Ma, Weiming Dong, and Changsheng Xu. 2023. Inversion-based style transfer with diffusion models. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 10146–10156

    2023

  33. [2015]

    ACM Trans

    Palette-based photo recoloring. ACM Trans. Graph. 34, 4 (2015), 139–1

    2015

  34. [2018]

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

    The unreasonable effectiveness of deep features as a perceptual metric. In Proceedings of the IEEE conference on computer vision and pattern recognition . 586–595

  35. [2021]

    Laion-400m: Open dataset of clip-filtered 400 million image-text pairs.arXiv preprint arXiv:2111.02114 (2021)

    2021 arXiv

  36. [2023]

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

    Segment anything. In Proceedings of the IEEE/CVF International Conference on Computer Vision. 4015–4026

  37. [2024]

    arXiv e-prints (2024), arXiv–2407

    ColorPeel: Color Prompt Learning with Diffusion Models via Color and Shape Disentanglement. arXiv e-prints (2024), arXiv–2407

    2024

Pith tools

Reviewed August 5, 2026 · model on record in the stance chip above.