Pith. sign in

REVIEW 4 major objections 5 minor 67 references

KinemaFX: A Kinematic-Driven Interactive System for Particle Effect Exploration and Customization

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read KinemaFX claims that non-expert users can find and compose particle effects by combining text with simple motion sketches, reducing mental effort.

desk verdict Solid, novel creative-support system whose core kinematic metric is asserted rather than validated; deserving of review but in need of honest statistical and perceptual-validation work. read the letter →

arxiv 2507.19782 v1 pith:YLA4LQAF submitted 2025-07-26 cs.HC

Yifei Zhang , Lin-Ping Yuan , Yuheng Zhao , Jielin Feng , Siming Chen This is my paper · ORCID
classification cs.HC
keywords particleeffectsinteractiveexplorationhuman-AIco-creationlargelanguagemodelskinematicrepresentationsimilaritysearchimplicitpreferenceguidanceuserstudy
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper tries to establish that non-expert users can explore and compose customized particle effect artworks without professional particle-system skills. KinemaFX lets users state intent as a weighted combination of a text description and a simple graphical sketch of emission shape, trail, and duration, then drives an iterative search that alternates between locally similar results and results along the direction of the user's implicit preference. Motion is represented as changes in spherical coordinates of an effect's boundary shape, combined with semantic embeddings into a structured representation. In a 16-participant within-subjects study against three ablated versions, the paper reports that the full system best supports intent expression, exploration, kinematic harmony, and reduced mental effort. The stake is whether a kinematic-first representation can make 3D dynamic visual effects as approachable as describing a scene in words and sketching a motion.

What carries the argument

The load-bearing object is the structured particle effect representation $R=(S,K)$, with $S$ a semantic embedding and $K=(\text{shape},\text{trail},\text{duration})$. Trail is encoded as a sequence of spherical-coordinate changes $(\Delta r_i,\Delta \theta_i,\Delta \phi_i)$ of the particles on the outer boundary of the initial emission shape across $N=8$ lifetime divisions, exploiting the effects' axial symmetry. The kinematic distance between two effects is computed per time step as the Hausdorff distance between transformed shapes plus a rotation penalty $\lambda(1-\cos\Delta\phi_j)$, summed over steps and multiplied by a duration adjustment factor; search minimizes this distance together with semantic cosine distance while allowing transformations such as translation, rotation, scaling, and duration adjustment.

What would settle it

Build a set of particle effect pairs, collect human similarity ratings of their motion, and compute the paper's kinematic distance $D_k$ using the same representation; if the rank correlation with human ratings is near zero, the metric fails and the search advantage claimed in the study would not generalize beyond the demo dataset.

Watch

Extended reading notes

Core claim

The central claim is that particle effects can be designed through kinematics as a first-class query language, with semantics as a flexible complement. The paper argues that because people describe motion poorly in words, the system should let them draw the emission shape, the trail, and the duration, and should let them pick promising intermediate results instead of rephrasing queries. It then claims that the kinematic behavior of many effects can be abstracted as the evolution of a primitive emission shape through time and encoded by the spherical-coordinate changes of boundary particles; this encoding, combined with semantic embeddings, supports search, alignment, and composition. The reported evaluation shows the full system outperforming ablations on intent expression, convergent and divergent exploration, kinematic harmony, and effort.

Load-bearing premise

The load-bearing premise is that the spherical-coordinate boundary-shape encoding, Hausdorff distance, and rotation penalty capture the motion features human viewers actually perceive; the paper offers illustrative examples but no human validation of this metric.

Editorial extensions

If this is right

  • Users who lack particle-system expertise can express motion intent through simple shapes and drawn lines, without first learning jargon or parameter panels.
  • By selecting satisfying intermediate results, users steer the search implicitly; the system alternates local and directional exploration to both converge on intent and avoid local optima.
  • Aligning kinematic behavior during retrieval reduces the manual repositioning, rescaling, and retiming normally needed to combine effects into one artwork.
  • Semantic and kinematic similarity can be weighted per query, so a user with a vague concept can lean on text, while a user with a clear motion in mind can lean on graphics.

Reading between the lines

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

  • The paper does not separately validate that its kinematic metric matches human perceptual similarity; a test of that mapping would show whether the study's benefits come from the metric or from the workflow around it.
  • Because the representation reduces motion to one composite shape, it may transfer to other multi-object dynamics, such as gesture or swarm motion, but only if the shape-abstraction step remains perceptually faithful in those domains.
  • The metric's weights $\lambda$ and $\alpha$ are tunable and unreported; a data-driven setting of these weights from human similarity ratings could make the method more general and easier to reproduce.
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

4 major / 5 minor

Summary. KinemaFX is an interactive system that supports non-expert users in exploring and customizing 3D particle effect artworks. The paper proposes a conceptual model separating semantic features from kinematic behavior, and builds a workflow around LLM-based semantic processing, graphical input of emission shapes/trails/duration, and an iterative exploration loop that alternates between local search around user-selected effects and directional extrapolation between two selected effects. The core technical contribution is a structured representation of particle effects (shape, trail, duration encoded in spherical coordinates) and a distance metric combining semantic cosine similarity with a kinematic distance based on Hausdorff shape distances, a rotation penalty, and a duration adjustment factor. The system is evaluated through two usage scenarios and a within-subjects user study with 16 participants comparing four conditions: Baseline, Preference-Guided Exploration, Kinematic-Driven Search, and full KinemaFX. The authors report that KinemaFX improves intent expression, supports convergent and divergent exploration, improves kinematic harmony, reduces mental effort, and leads to satisfying effect artworks.

Significance. If the reported results hold, KinemaFX would be a useful step toward making complex 3D particle-effect creation accessible to non-experts, and its abstraction of kinematic behavior as transformations of a composite shape is a reasonable design idea that could generalize to other dynamic effects. The paper has clear strengths: a formative study with interviews and dataset analysis that motivates the design goals, a concrete and well-integrated system description, a user study with counterbalanced within-subjects conditions, and qualitative quotes that give insight into user experience. The authors also explicitly acknowledge some limitations, such as the lack of independent LLM evaluation. However, the quantitative evidence for the central claims is currently under-reported, and the kinematic distance metric—the paper's main technical novelty—is not validated against human perception and depends on unreported free parameters. These issues are load-bearing because the user study is the only empirical support for the claimed benefits of kinematic-driven search.

major comments (4)
  1. [Section 5.2, Figure 11] The quantitative basis for the central claim is under-reported. The text states that repeated-measures ANOVA or Friedman tests with Bonferroni correction were used, but it does not report F or chi-square statistics, degrees of freedom, p-values, means, standard deviations, or effect sizes for any of Q1–Q15. With n=16 and six pairwise comparisons per item, a bare significance star in Figure 11 is not sufficient to determine which differences actually reached the corrected threshold, and the reader cannot assess the magnitude of the reported advantages. Please add a full statistics table with descriptive statistics, test statistics, and effect sizes, and clarify in the figure which comparisons each star denotes.
  2. [Section 4.3.3, Eqs. (5)–(8)] The distance metric D_k depends on the free parameters lambda (rotation penalty weight) and alpha (duration adjustment weight), but the manuscript never reports these values, their ranges, or a sensitivity analysis. Since the rankings produced by Eq. (4) are directly controlled by these weights, the search component is not reproducible as described, and the robustness of the claimed benefits of kinematic-driven search cannot be assessed. Please report the parameter settings used in the study and include a sensitivity analysis, or otherwise justify the chosen values.
  3. [Section 5.1, study conditions] The ablation does not isolate the contribution of the kinematic distance metric. The Kinematic-Driven Search condition introduces both graphical kinematic input and the metric D_k relative to Baseline, while KinemaFX adds preference-guided exploration on top; consequently, the differences attributed to kinematic-driven search could be produced by the additional input modality alone, with no evidence that D_k itself aligns with human perception. Please add a control condition that provides graphical input while ranking by semantic similarity only, or provide an independent validation of D_k against human similarity judgments (for example, a perceptual rating study over retrieved rankings), and temper the claims accordingly.
  4. [Section 3.3 and Section 2.1] The statement that the proposed kinematic representation "better aligns with human visual perception" is asserted rather than demonstrated. The qualitative analysis of the dataset establishes that axial symmetry is common and that within-artwork kinematic features are more similar than between-artwork features, but it does not test whether the spherical-coordinate trail encoding or the Hausdorff-plus-rotation-penalty distance matches human judgments of kinematic similarity. Figure 8 shows illustrative examples, not a validation. Given that the kinematic metric is the paper's core technical contribution, this missing validation is load-bearing for the generality of the search claims.
minor comments (5)
  1. [Figure 11] The significance stars are not accompanied by a description of which pairwise comparisons are shown; please add brackets, annotations, or a table so the reader can determine which conditions are being compared.
  2. [Abstract and Contributions] The phrase "customarily creating particle effect artworks" appears to be a typo; "in a customized manner" or "customized creation" would be clearer.
  3. [References] Reference [40] appears malformed after "Lars St"; the author names should be corrected.
  4. [Sections 4.2.2 and 4.3.1] The default values N=8 for lifetime discretization and K=4 for the number of local results are stated without justification; a brief explanation or a pilot-test reference would improve reproducibility.
  5. [Section 3.3] The comparison of within-artwork versus between-artwork feature differences is reported only with a significance threshold; please provide the test used and the sample sizes for that analysis.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the kinematic metric's unvalidated perceptual claims are an empirical validity concern, not a derivation-loop.

full rationale

The paper's central effectiveness claim is supported by a within-subjects user study with ablation conditions, Likert ratings, and qualitative interviews, not by quantities derived from its own kinematic equations. The kinematic distance metric in Eqs. 5-8 has unreported weights lambda and alpha and is not validated against human perceptual judgments, but this is a correctness or generalizability risk, not circularity: no parameter is fitted to the study outcomes, and no reported result is equal to an input by construction. The kinematic representation was informed by the same collected asset set used in the study, yet the evaluation measures subjective experience and task outcomes rather than the metric's ranking accuracy, so there is no fitted-input-called-prediction loop. Self-citations to prior work by the same authors (e.g., refs. 51, 57, 58, 59) appear only in related-work context and instrumentation references, and none carries the load of the paper's technical claims. No uniqueness theorem, no imported ansatz, and no renamed empirical pattern is used to force the system's design. Therefore, no specific circular step can be exhibited under the required standard.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the perceptual adequacy of a hand-designed kinematic representation, on unstated metric weights, and on a proprietary dataset. No new physical entities are postulated. The effectiveness claim is supported by a user study, but the representation itself is not independently validated.

free parameters (4)
  • lambda (rotation penalty weight) = not reported
    Introduced in Eq. 5 to balance Hausdorff shape distance against rotation difference; no value or tuning procedure is given, yet it directly controls kinematic distance and search ranking.
  • alpha (duration adjustment weight) = not reported
    In Eq. 7, alpha controls how strongly duration mismatch amplifies trail distance; no value or fitting is reported.
  • N (lifetime discretization) = 8
    Default number of time segments in Eq. 3 and trail distance summation; chosen without justification.
  • K (top results per local search) = 4
    Default number of recommended effects in local exploration; set by hand.
assumptions (5)
  • domain assumption Particle effect emission shapes are adequately represented by circle, cylinder, and sphere primitives.
    Sections 3.3 and 4.3.1; point shapes are collapsed into the other primitives, and non-axial cases are excluded.
  • domain assumption Particle effects are approximately axisymmetric, allowing alignment with the polar axis.
    Justified in Section 3.3 by 97.4% axial symmetry, but the remaining cases and representational choices still limit coverage; used in Eq. 3.
  • ad hoc to paper Hausdorff distance on transformed primitive shapes plus rotation penalty captures perceptually meaningful kinematic similarity.
    Eq. 5 is proposed without calibration against human similarity judgments; the central search behavior depends on this assumption.
  • domain assumption LLM semantic standardization and candidate generation preserve user intent well enough for retrieval and directional exploration.
    The paper notes kinematic details are poorly conveyed by LLMs in Section 6.3 and does not evaluate LLM accuracy or compare models.
  • domain assumption The curated dataset of 147 artworks and 839 effects from two Unity Asset Store studios is representative of diverse particle effects.
    Section 3.1; this constrains the generality of the kinematic analysis and the user study.

how reviews work

0 comments
Cite this review

Pith. "Pith review of KinemaFX: A Kinematic-Driven Interactive System for Particle Effect Exploration and Customization." pith.science (2026). https://pith.science/paper/YLA4LQAF

@misc{pith2026250719782,
  author       = {Pith},
  title        = {Pith review of: KinemaFX: A Kinematic-Driven Interactive System for Particle Effect Exploration and Customization},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YLA4LQAF}},
  note         = {Machine review of arXiv:2507.19782}
}
read the original abstract

Particle effects are widely used in games and animation to simulate natural phenomena or stylized visual effects. However, creating effect artworks is challenging for non-expert users due to their lack of specialized skills, particularly in finding particle effects with kinematic behaviors that match their intent. To address these issues, we present KinemaFX, a kinematic-driven interactive system, to assist non-expert users in constructing customized particle effect artworks. We propose a conceptual model of particle effects that captures both semantic features and kinematic behaviors. Based on the model, KinemaFX adopts a workflow powered by Large Language Models (LLMs) that supports intent expression through combined semantic and kinematic inputs, while enabling implicit preference-guided exploration and subsequent creation of customized particle effect artworks based on exploration results. Additionally, we developed a kinematic-driven method to facilitate efficient interactive particle effect search within KinemaFX via structured representation and measurement of particle effects. To evaluate KinemaFX, we illustrate usage scenarios and conduct a user study employing an ablation approach. Evaluation results demonstrate that KinemaFX effectively supports users in efficiently and customarily creating particle effect artworks.

Figures

Figures reproduced from arXiv: 2507.19782 by the authors.

Figure 1
Figure 1. An overview of KinemaFX. KinemaFX supports interactive particle effect exploration and customization through [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Effect artworks from the collected dataset. Our [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. The distribution of particle effects across key dimensions. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Common cases of particle effects in terms of emis [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Examples of graphical input. Users can intuitively [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: This case demonstrates the interplay between local and directional exploration: In Step 1, the user begins with a [PITH_FULL_IMAGE:figures/full_fig_p009_6.png]
Figure 7
Figure 7. Figure 7: We propose a two-stage kinematic-driven method to enhance particle effect search and alignment: convert constraints [PITH_FULL_IMAGE:figures/full_fig_p010_7.png]
Figure 8
Figure 8. Figure 8: Example particle effects with different kinematic [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: We demonstrate the usage of KinemaFX through two usage scenarios. (a) “Charge-explosion” effect: The user divided [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: Representative effect artworks created by users un [PITH_FULL_IMAGE:figures/full_fig_p013_10.png]
Figure 11
Figure 11. Figure 11: Quantitative results of the user study. Significance levels are denoted as follows: (*) for [PITH_FULL_IMAGE:figures/full_fig_p014_11.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

67 extracted references · 54 canonical work pages

  1. [1]

    Matias Acuña. 2024. Visual Effects for Stilized Video Games. (2024)

    work page 2024

  2. [2]

    Shm Garanganao Almeda, JD Zamfirescu-Pereira, Kyu Won Kim, Pradeep Mani Rathnam, and Bjoern Hartmann. 2024. Prompting for Discovery: Flex- ible Sense-Making for AI Art-Making with Dreamsheets. In Proceedings of the CHI Conference on Human Factors in Computing Systems . 1–17

    2024

  3. [3]

    Milton Friedman and. 1937. The Use of Ranks to Avoid the Assumption of Normality Implicit in the Analysis of Variance. J. Amer. Statist. Assoc. 32, 200 (1937), 675–701. arXiv:https://www.tandfonline.com/doi/pdf/10.1080/01621459.1937.10503522 doi:10.1080/01621459.1937.10503522

    arXiv 1937

  4. [4]

    Tyler Angert, Miroslav Suzara, Jenny Han, Christopher Pondoc, and Hariharan Subramonyam. 2023. Spellburst: A node-based interface for exploratory creative coding with natural language prompts. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology . 1–22

    work page 2023

  5. [5]

    Rahul Arora, Rubaiat Habib Kazi, Danny M Kaufman, Wilmot Li, and Karan Singh

  6. [6]

    Liuqing Chen, Qianzhi Jing, Yixin Tsang, Qianyi Wang, Ruocong Liu, Duowei Xia, Yunzhan Zhou, and Lingyun Sun. 2024. AutoSpark: Supporting Automobile Appearance Design Ideation with Kansei Engineering and Generative AI. In Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology. 1–19

    work page 2024

  7. [7]

    Chia-Hsing Chiu, Yuki Koyama, Yu-Chi Lai, Takeo Igarashi, and Yonghao Yue

    work page

  8. [8]

    Byungkuk Choi, Roger Blanco i Ribera, John P Lewis, Yeongho Seol, Seokpyo Hong, Haegwang Eom, Sunjin Jung, and Junyong Noh. 2016. SketchiMo: sketch- based motion editing for articulated characters. ACM Transactions on Graphics (ToG) 35, 4 (2016), 1–12

    work page 2016

Show all 67 references
  1. [9]

    DaEun Choi, Sumin Hong, Jeongeon Park, John Joon Young Chung, and Juho Kim. 2024. CreativeConnect: Supporting Reference Recombination for Graphic Design Ideation with Generative AI. In Proceedings of the CHI Conference on Human Factors in Computing Systems . 1–25

    2024

  2. [10]

    Yunjey Choi, Minje Choi, Munyoung Kim, Jung-Woo Ha, Sunghun Kim, and Jaegul Choo. 2018. Stargan: Unified generative adversarial networks for multi- domain image-to-image translation. In Proceedings of the IEEE conference on computer vision and pattern recognition . 8789–8797

    2018

  3. [11]

    Jorge Chueca, Carlos Cetina, Oscar Pastor, and Jaime Font. 2024. Search-based co-creation of software models: The case of particle systems for video games. Information and Software Technology 171 (2024), 107466

    2024

  4. [12]

    John Joon Young Chung and Eytan Adar. 2023. Artinter: AI-Powered Boundary Objects for Commissioning Visual Arts. InProceedings of the 2023 ACM Designing Interactive Systems Conference. 1997–2018

    2023

  5. [13]

    John Joon Young Chung and Eytan Adar. 2023. Promptpaint: Steering text-to- image generation through paint medium-like interactions. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology . 1–17

    2023

  6. [14]

    Hai Dang, Frederik Brudy, George Fitzmaurice, and Fraser Anderson. 2023. World- Smith: Iterative and Expressive Prompting for World Building with a Generative AI. In Proceedings of the 36th Annual ACM Symposium on User Interface Software and Technology. 1–17

    2023

  7. [15]

    Charles Finance and Susan Zwerman. 2015. The visual effects producer: under- standing the art and business of VFX . Routledge

    2015

  8. [16]

    Rao Fu, Zehao Wen, Zichen Liu, and Srinath Sridhar. 2024. Anyhome: Open- vocabulary generation of structured and textured 3d homes. In European Confer- ence on Computer Vision . Springer, 52–70

    2024

  9. [17]

    Epic Games. 2023. Cascade Particle System. https://dev.epicgames. com/documentation/en-us/unreal-engine/cascade-particle-editor- reference?application_version=4.27

    2023

  10. [18]

    Epic Games. 2023. Niagara Particle System. https://dev.epicgames.com/ documentation/en-us/unreal-engine/overview-of-niagara-effects-for-unreal- engine

    2023

  11. [19]

    Martin Guay, Rémi Ronfard, Michael Gleicher, and Marie-Paule Cani. 2015. Space- time sketching of character animation. ACM Transactions on Graphics (ToG) 34, 4 (2015), 1–10

    2015

  12. [20]

    Sandra G Hart and Lowell E Staveland. 1988. Development of NASA-TLX (Task Load Index): Results of empirical and theoretical research. In Advances in psy- chology. Vol. 52. Elsevier, 139–183

    1988

  13. [21]

    Erin J Hastings, Ratan K Guha, and Kenneth O Stanley. 2008. Interactive evolution of particle systems for computer graphics and animation. IEEE Transactions on Evolutionary Computation 13, 2 (2008), 418–432

    2008

  14. [22]

    Rong Huang, Haichuan Lin, Chuanzhang Chen, Kang Zhang, and Wei Zeng. 2024. PlantoGraphy: Incorporating iterative design process into generative artificial intelligence for landscape rendering. In Proceedings of the CHI Conference on Human Factors in Computing Systems . 1–19

    2024

  15. [23]

    VES Jeffrey Okun and VES Susan Zwerman. 2020. The VES handbook of visual effects: industry standard VFX practices and procedures . Routledge

    2020

  16. [24]

    Janin Koch, Nicolas Taffin, Michel Beaudouin-Lafon, Markku Laine, Andrés Lucero, and Wendy E Mackay. 2020. Imagesense: An intelligent collaborative ideation tool to support diverse human-computer partnerships. Proceedings of the ACM on human-computer interaction 4, CSCW1 (2020), 1–27

    2020

  17. [25]

    Yuki Koyama and Masataka Goto. 2022. BO as assistant: Using Bayesian opti- mization for asynchronously generating design suggestions. In Proceedings of the 35th Annual ACM Symposium on User Interface Software and Technology . 1–14

    2022

  18. [26]

    Boyu Li, Linping Yuan, Zhe Yan, Qianxi Liu, Yulin Shen, and Zeyu Wang. 2024. Anicraft: Crafting everyday objects as physical proxies for prototyping 3d char- acter animation in mixed reality. In Proceedings of the Annual ACM Symposium on User Interface Software and Technology . 1–14

    2024

  19. [27]

    Jiawei Lin, Jiaqi Guo, Shizhao Sun, Zijiang Yang, Jian-Guang Lou, and Dongmei Zhang. 2023. Layoutprompter: awaken the design ability of large language models. Advances in Neural Information Processing Systems 36 (2023), 43852–43879

    2023

  20. [28]

    Vivian Liu, Rubaiat Habib Kazi, Li-Yi Wei, Matthew Fisher, Timothy Langlois, Seth Walker, and Lydia Chilton. 2025. Logomotion: Visually-grounded code synthesis for creating and editing animation. In Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems . 1–16

    2025

  21. [29]

    Vivian Liu, Jo Vermeulen, George Fitzmaurice, and Justin Matejka. 2023. 3DALL- E: Integrating text-to-image AI in 3D design workflows. In Proceedings of the 2023 ACM designing interactive systems conference . 1955–1977

    2023

  22. [30]

    Aitor Luque Bodet. 2022. Node-based particle system . B.S. thesis. Universitat Politècnica de Catalunya

    2022

  23. [31]

    Damien Masson, Sylvain Malacria, Géry Casiez, and Daniel Vogel. 2024. Direct- gpt: A direct manipulation interface to interact with large language models. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems . 1–16

    2024

  24. [32]

    Meri Mattila. 2018. Visual effects for a 3D action game. (2018)

    2018

  25. [33]

    We Are Visual Thinkers, Not Verbal Thinkers!

    Hyerim Park, Joscha Eirich, Andre Luckow, and Michael Sedlmair. 2024. " We Are Visual Thinkers, Not Verbal Thinkers!": A Thematic Analysis of How Professional Designers Use Generative AI Image Generation Tools. In Proceedings of the 13th Nordic Conference on Human-Computer Int...

    2024

  26. [34]

    Xiaohan Peng, Janin Koch, and Wendy E Mackay. 2024. Designprompt: Using multimodal interaction for design exploration with generative ai. In Proceedings of the 2024 ACM Designing Interactive Systems Conference . 804–818

    2024

  27. [35]

    Wanli Qian, Chenfeng Gao, Anup Sathya, Ryo Suzuki, and Ken Nakagaki. 2024. SHAPE-IT: Exploring Text-to-Shape-Display for Generative Shape-Changing Behaviors with LLMs. In Proceedings of the 37th Annual ACM Symposium on User Interface Software and Technology. 1–29

    2024

  28. [36]

    Joona Raappana. 2018. Great Particles and How to Make Them: Development of guidelines for the Unity Particle System. (2018)

    2018

  29. [37]

    Rintar¯o Sat¯o and Michael Cohen. 2022. Particle System Parametrization with Bimanual Hand Gestures. In SHS Web of Conferences , Vol. 139. EDP Sciences, 03023

    2022

  30. [38]

    Xinyu Shi, Yinghou Wang, Ryan Rossi, and Jian Zhao. 2025. Brickify: Enabling Expressive Design Intent Specification through Direct Manipulation on Design Tokens. In Proceedings of the CHI Conference on Human Factors in Computing Systems. 1–20

    2025

  31. [39]

    Kihoon Son, DaEun Choi, Tae Soo Kim, Young-Ho Kim, and Juho Kim. 2024. Gen- query: Supporting expressive visual search with generative models. InProceedings of the CHI Conference on Human Factors in Computing Systems . 1–19. KinemaFX: A Kinematic-Driven Interactive System for ...

    2024

  32. [40]

    Lars St, Svante Wold, et al. 1989. Analysis of variance (ANOVA). Chemometrics and intelligent laboratory systems 6, 4 (1989), 259–272

    1989

  33. [41]

    Sangho Suh, Meng Chen, Bryan Min, Toby Jia-Jun Li, and Haijun Xia. 2024. Luminate: Structured Generation and Exploration of Design Space with Large Language Models for Human-AI Co-Creation. InProceedings of the CHI Conference on Human Factors in Computing Systems . 1–26

    2024

  34. [42]

    Hou In Ivan Tam, Hou In Derek Pun, Austin T Wang, Angel X Chang, and Manolis Savva. 2024. SceneMotifCoder: Example-driven visual program learning for generating 3D object arrangements. arXiv preprint arXiv:2408.02211 (2024)

    2024 arXiv

  35. [43]

    Unity Technologies. [n. d.]. Visual Effect Graph. https://unity.com/features/visual- effect-graph

  36. [44]

    Unity Technologies. 2023. Shuriken Particle System. https://docs.unity3d.com/ 355/Documentation/Components/class-ParticleSystem.html

    2023

  37. [45]

    Tiffany Tseng, Ruijia Cheng, and Jeffrey Nichols. 2024. Keyframer: Empowering animation design using large language models. arXiv preprint arXiv:2402.06071 (2024)

    2024 arXiv

  38. [46]

    Tero Vääräniemi. 2016. Example Solutions of Visual Effects in Fantasy Games. (2016)

    2016

  39. [47]

    Cheng Wang, Haojin Yang, and Christoph Meinel. 2015. Deep semantic mapping for cross-modal retrieval. In 2015 IEEE 27th International conference on tools with artificial intelligence (ICTAI). IEEE, 234–241

    2015

  40. [48]

    Can Wang, Hongliang Zhong, Menglei Chai, Mingming He, Dongdong Chen, and Jing Liao. 2024. Chat2Layout: Interactive 3D furniture layout with a multimodal LLM. arXiv preprint arXiv:2407.21333 (2024)

    2024 arXiv

  41. [49]

    Shun-Yu Wang, Wei-Chung Su, Serena Chen, Ching-Yi Tsai, Marta Misztal, Kather- ine M Cheng, Alwena Lin, Yu Chen, and Mike Y Chen. 2024. RoomDreaming: Generative-AI Approach to Facilitating Iterative, Preliminary Interior Design Exploration. In Proceedings of the CHI Conference...

    2024

  42. [50]

    Zhijie Wang, Yuheng Huang, Da Song, Lei Ma, and Tianyi Zhang. 2024. PromptCharm: Text-to-Image Generation through Multi-modal Prompting and Refinement. In Proceedings of the CHI Conference on Human Factors in Computing Systems. 1–21

    2024

  43. [51]

    Zhan Wang, Lin-Ping Yuan, Liangwei Wang, Bingchuan Jiang, and Wei Zeng

  44. [52]

    Eric W Weisstein. 2004. Bonferroni correction. https://mathworld. wolfram. com/ (2004)

    2004

  45. [53]

    Shishi Xiao, Liangwei Wang, Xiaojuan Ma, and Wei Zeng. 2024. TypeDance: Creating semantic typographic logos from image through personalized genera- tion. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems. 1–18

    2024

  46. [54]

    Jun Xing, Rubaiat Habib Kazi, Tovi Grossman, Li-Yi Wei, Jos Stam, and George Fitzmaurice. 2016. Energy-brushes: Interactive tools for illustrating stylized elemental dynamics. InProceedings of the 29th Annual Symposium on User Interface Software and Technology. 755–766

    2016

  47. [55]

    Hui Ye, Kin Chung Kwan, Wanchao Su, and Hongbo Fu. 2020. ARAnimator: In-situ character animation in mobile AR with user-defined motion gestures. ACM Transactions on Graphics (TOG) 39, 4 (2020), 83–1

    2020

  48. [56]

    Emilie Yu, Kevin Blackburn-Matzen, Cuong Nguyen, Oliver Wang, Rubaiat Habib Kazi, and Adrien Bousseau. 2023. Videodoodles: Hand-drawn anima- tions on videos with scene-aware canvases. ACM Transactions on Graphics (TOG) 42, 4 (2023), 1–12

    2023

  49. [57]

    Lin-Ping Yuan, John J Dudley, Per Ola Kristensson, and Huamin Qu. 2025. Per- sonalized Dual-Level Color Grading for 360-degree Images in Virtual Reality. IEEE Transactions on Visualization and Computer Graphics (2025)

    2025

  50. [58]

    You’ll Be Alice Adventuring in Wonderland!

    Lin-Ping Yuan, Feilin Han, Liwenhan Xie, Junjie Zhang, Jian Zhao, and Huamin Qu. 2025. " You’ll Be Alice Adventuring in Wonderland!" Processes, Challenges, and Opportunities of Creating Animated Virtual Reality Stories. In Proceedings of the CHI Conference on Human Factors in ...

    2025

  51. [59]

    Lin-Ping Yuan, Ziqi Zhou, Jian Zhao, Yiqiu Guo, Fan Du, and Huamin Qu. 2021. Infocolorizer: Interactive recommendation of color palettes for infographics.IEEE Transactions on Visualization and Computer Graphics 28, 12 (2021), 4252–4266

    2021

  52. [60]

    Xingchen Zeng, Ziyao Gao, Yilin Ye, and Wei Zeng. 2024. IntentTuner: an interactive framework for integrating human intentions in fine-tuning text-to- image generative models. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems . 1–18

    2024

  53. [61]

    Bingqing Zhang and Wenfeng Hu. 2017. Game special effect simulation based on particle system of Unity3D. In 2017 IEEE/ACIS 16th International Conference on Computer and Information Science (ICIS) . IEEE, 595–598

    2017

  54. [62]

    Enhao Zhang and Nikola Banovic. 2021. Method for exploring generative adver- sarial networks (gans) via automatically generated image galleries. InProceedings of the 2021 CHI Conference on Human Factors in Computing Systems . 1–15

    2021

  55. [63]

    Hao Zhou, Chengming Hu, Ye Yuan, Yufei Cui, Yili Jin, Can Chen, Haolun Wu, Dun Yuan, Li Jiang, Di Wu, et al. 2024. Large language model (llm) for telecom- munications: A comprehensive survey on principles, key techniques, and oppor- tunities. IEEE Communications Surveys & Tuto...

    2024

  56. [64]

    Qian Zhou, David Ledo, George Fitzmaurice, and Fraser Anderson. 2024. TimeTun- nel: Integrating Spatial and Temporal Motion Editing for Character Animation in Virtual Reality. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems. 1–17

    2024

  57. [2019]

    In Proceedings of the 32nd annual ACM symposium on user interface software and technology

    Magicalhands: Mid-air hand gestures for animating in vr. In Proceedings of the 32nd annual ACM symposium on user interface software and technology . 463–477

  58. [2020]

    ACM Transactions on Graphics (TOG) 39, 4 (2020), 85–1

    Human-in-the-loop differential subspace search in high-dimensional latent space. ACM Transactions on Graphics (TOG) 39, 4 (2020), 85–1

    2020

  59. [2024]

    In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems

    Virtuwander: Enhancing multi-modal interaction for virtual tour guidance through large language models. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems . 1–20

    2024

Pith tools

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