Pith. sign in

REVIEW 3 major objections 5 minor 62 references

SketchConcept: Sketching-based Concept Recomposition for Product Design using Generative AI

T0 review · 3 major / 5 minor · reviewed 2026-08-05 · deepseek-v4-flash

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

Pith's one-line read A pipeline that decomposes AI-generated product images into editable functional parts, letting designers change one component without regenerating the entire concept.

desk verdict A genuinely novel pipeline for component-level design exploration, but the core claim that edits are component-local is not yet supported; worth a serious referee and a requested revision. read the letter →

arxiv 2508.07141 v1 pith:DJR3V3NR submitted 2025-08-10 cs.HC

Runlin Duan , Chenfei Zhu , Yuzhao Chen , Dizhi Ma , Jingyu Shi , Ziyi Liu , Karthik Ramani This is my paper
classification cs.HC
keywords DesignSupportSystemSketchandTextInputUserExperienceGenerativeAIConceptDecompositionSemanticSegmentationFunction-to-VisualMappingComponent-levelEditing
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

SketchConcept claims that a text-described function—say, 'headlights: slim'—can be tied to a specific segmented region of a generated concept image, and that this function-to-visual mapping lets a designer swap or redraw that single component while leaving the rest of the design intact. The paper builds the full loop: sketch plus voice input refined by GPT-4V, concept images from DALL-E 3, semantic segmentation into components, GPT-4V assigning functions to colored regions, and DALL-E 2 applying local edits triggered by function-chart selections or re-sketching. The authors validate the parts with a 600-image annotated dataset (mean IoU up to 0.951 across segmentation models) and a ten-user study, and report that component-level editing works but off-the-shelf models sometimes misread intent, as when a Nerf Gun sketch was turned into a firing cannon. If correct, the system gives generative design a handle on a long-standing pain point: 'change only this part' rather than 'regenerate everything and hope.'

What carries the argument

The function-to-visual mapping workflow: (1) a custom-trained DeepLabV3 semantic segmentation model splits the generated concept image into components; (2) GPT-4V, shown the image with components color-overlaid, assigns each text-described function to a segment; (3) a per-component function chart is generated; and (4) editing—via selecting an alternative or re-sketching—is executed by DALL-E 2 on that segment alone. The mapping is what converts an unstructured image from a text-to-image model into a structure that supports local, semantic editing.

What would settle it

A controlled study where the same concept image is edited on one component (e.g., headlights from slim to circular) and the non-target segments are compared pixel-wise before and after; if the non-target regions change beyond a small tolerance, the component-level editing claim is false. A simpler observable already in the paper: the user quote 'it keeps changing my sketch to a firing cannon instead of a Nerf Gun' shows the assumption can break; a systematic count of such failures across many edits would settle how often the pipeline truly preserves the rest of the design.

Watch

Extended reading notes

Core claim

The central discovery claimed by SketchConcept is that function and form can be aligned at the component level inside a generative pipeline. Given a single AI-generated concept image, the system overlays each semantic segment with a distinct color and asks GPT-4V which color region corresponds to each function extracted from the design description; the result is a function chart that offers alternative solutions (e.g., headlight: slim → rectangle/circle). Selecting an alternative updates only the associated segment through DALL-E 2, and redrawing the component does the same through the same refinement path used at generation. The paper's own measurements put semantic segmentation at 0.896–0.

Load-bearing premise

The system assumes that the off-the-shelf GPT-4V and DALL-E models will, on request, map a function to the exact right pixel segment and then regenerate that segment alone without altering the rest of the image.

Editorial extensions

If this is right

  • Component-level exploration becomes possible: a designer can evaluate alternative implementations of one function (e.g., trunk style, headlight shape) while other parts of the concept stay fixed.
  • The dataset and evaluation—600 annotated DALL-E 3 concept images, six segmentation architectures, function-mapping accuracy per category—provide a benchmark for measuring how reliably GenAI models can link function to visual part.
  • The same decomposition pipeline can be retargeted to other concept-design domains, such as interior layout, where existing semantic segmentation datasets can be reused.
  • Users can combine AI suggestions with their own sketches: when the recommended alternatives don't match intent, the user redraws the masked component and the system regenerates only that region.
  • The system's utility as an idea-generation aid was recognized in the user study, with participants noting the component explorer surfaced options they hadn't thought of.

Reading between the lines

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

  • If the mapping holds up across a wider range of products, the same decomposition could serve as a data-generation tool: automatically producing labeled (function, part) pairs from unlabeled GenAI images, which would let future systems be trained with far less manual annotation.
  • A natural stress test is whether segmentation masks remain stable after an edit; if DALL-E 2's local regeneration shifts boundaries, the next edit may target the wrong region. Measuring mask IoU before and after each edit would quantify this drift.
  • The approach is currently locked to the appearance of the 2D image; extending function-to-visual mapping to 3D or interactive prototypes would require a different representation of 'component' that survives regeneration.
  • Its dependence on off-the-shelf GPT-4V and DALL-E means the claim is really about what these current models can do together, not about a novel generative model; if those APIs change behavior, the demonstrated accuracy numbers are not guaranteed.
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

3 major / 5 minor

Summary. SketchConcept is an end-to-end system that takes a freehand sketch and voice description, uses GPT-4V to refine the description and DALL-E 3 to generate concept images, then semantically segments the image with a fine-tuned DeepLabV3 model and uses GPT-4V to map user functions to visual components. The system presents a function chart per component and supports editing either by selecting an alternative solution (which prompts DALL-E 2 with 'change function from A to B') or by sketching the component on a white canvas. The authors contribute a labeled dataset of 600 DALL-E 3 images across three product categories, compare six segmentation models (mIoU 0.896–0.951), evaluate GPT-4V function-mapping accuracy (80.7–100%), and report a user study with ten mechanical engineering undergraduates.

Significance. If the component-level editing were reliable, SketchConcept would be a valuable addition to conceptual design tools, integrating functional and visual decomposition in a single workflow. The released dataset and the segmentation/function-mapping benchmarks are concrete assets, and the paper is commendable for reporting quantitative results on those modules rather than showing demos only. However, the paper's central novelty—localized, component-level modification of a generated concept—rests on an evaluation that is currently too weak: no spatial-locality metric, no baseline condition, a small homogeneous user group, and admitted failure cases in the paper itself.

major comments (3)
  1. [§3.5.1, §5.4, §7.3] The edit pipeline passes only a text prompt to DALL-E 2 ("change function from [SOLUTION_A] to [SOLUTION_B]"), with no mask, bounding box, or other spatial constraint. No experiment measures whether non-target components remain intact after an edit. The only support is the subjective Likert item Q7 (avg=4.1, std=1.1) and selected user quotes. Section 7.3 itself documents failures where the whole concept changes (a Nerf Gun becomes a firing cannon; a magazine orientation is flipped), which is consistent with full-image regeneration rather than localized editing. I recommend adding a quantitative edit-locality metric (e.g., IoU or pixel difference outside the target component's mask) and reporting it across edits, and/or comparing against whole-image regeneration.
  2. [§6] The user study has only 10 participants from a single discipline (mechanical engineering undergraduates), no control or baseline condition, and only Likert-scale self-report data. The claim that SketchConcept supports 'seamless iteration and adaptation' (DG3) is not supported beyond anecdotal quotes. A comparative study (e.g., SketchConcept vs. prompt-only DALL-E editing) with a more diverse sample, and reporting of effect sizes, would be needed to substantiate the system's specific benefit.
  3. [§6.1] The results report Q3 twice with different values: "Q3: avg=3.4, std=0.97" and "Q3: avg=3.0, std=1.1". Based on Figure 6, the second instance appears to be Q4 ('I am satisfied with the image generated by the system'). This inconsistency makes the descriptive statistics unreliable and must be corrected.
minor comments (5)
  1. [§5.2] 'valuation set' should be 'validation set'.
  2. [§3.5.1] Spelling of 'DALLE-2' vs 'DALL-E 2' is inconsistent; use the latter throughout.
  3. [Figure 2] The caption contains 'as illustrated in ??' with a missing cross-reference.
  4. [Abstract & §6] The 'two-session user study' is not described; only the second session is mentioned in Figure 7. Clarify the session structure.
  5. [Table 2] The heading says 'Qualitative results' but the table reports quantitative accuracy percentages; rename to 'Function mapping accuracy' or similar.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's central claims are supported by held-out segmentation and function-mapping benchmarks and by user studies, not by fitting inputs to predictions.

full rationale

SketchConcept is an empirical systems/HCI paper. Its substantive claims are supported by three independent forms of evidence. First, the semantic segmentation component is trained on an 80% split of a newly collected dataset and evaluated on a disjoint 10% test split (Section 5.2, Table 1), so the reported mean IoU is a held-out benchmark rather than a fitted-input prediction. Second, the function-to-visual mapping claim is tested by asking GPT-4V to classify a fixed set of functions against the same dataset's images (Section 5.3, Table 2), again on held-out data with pre-specified function labels; this is an evaluation of the actual deployed component on unseen examples, not a quantity forced by construction. Third, the component-level modification claim is assessed through a two-session user study with Likert ratings, SUS, and interviews (Section 6), which is a subjective usability evaluation rather than a derivation from inputs. The paper does use GPT-4V both as the system's mapping module and as the subject of the mapping accuracy evaluation, and it uses DALL-E both to generate the dataset and to perform edits, but this overlap is not circularity: the reported numbers are independent judgments on fixed examples. The concern in Section 7.3 that DALL-E sometimes fails to preserve spatial intent (e.g., the Nerf Gun becoming a firing cannon) is a validity/robustness limitation, not a circularity. No load-bearing claim reduces to its own inputs, no fitted parameter is renamed as a prediction, and no argument depends on a self-citation or an imported uniqueness theorem. Score 0.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No invented physical or mathematical entities are introduced; the paper is a systems integration over off-the-shelf APIs and standard segmentation models. The central claim rests on empirical assumptions about the reliability of commercial generative models and the representativeness of a small self-collected dataset.

free parameters (1)
  • per-category component taxonomy = Car: 5 components, Nerf Gun: 3, Robot Dog: 3
    The granularity of decomposition (which regions are separate components) is chosen by hand in Section 5.1 and defines both the segmentation labels and the function charts; a different taxonomy would change the evaluation numbers and user experience.
assumptions (4)
  • domain assumption GPT-4V can reliably map a text function description to the correct color-coded segmented region of a generated concept image.
    This is the core of the function-to-visual mapping; evaluated only on 6-10 fixed functions per category in Section 5.3, and only on DALL-E-generated images from the authors' own dataset.
  • domain assumption DALL-E 3 and DALL-E 2 can generate concept images and perform component-level edits from refined text prompts without distorting the rest of the image.
    Generation in Section 3.3 and editing in Section 3.5 both rely on commercial text-to-image APIs; Section 7.3 reports failures where the model ignored the user's sketch (e.g., the Nerf Gun became a cannon).
  • domain assumption The 600-image dataset of DALL-E-generated car, Nerf Gun, and Robot Dog images with the authors' component annotations is representative of conceptual product design images.
    Section 5.1 collects only these three categories and only isometric, clean-background images; the segmentation and mapping results may not transfer to other products or styles.
  • domain assumption Ten mechanical engineering undergraduates (ages 19-23) are a valid proxy for conceptual product designers.
    Section 6 draws conclusions about design support efficacy from 10 non-professional participants with no baseline condition.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SketchConcept: Sketching-based Concept Recomposition for Product Design using Generative AI." pith.science (2026). https://pith.science/paper/DJR3V3NR

@misc{pith2026250807141,
  author       = {Pith},
  title        = {Pith review of: SketchConcept: Sketching-based Concept Recomposition for Product Design using Generative AI},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DJR3V3NR}},
  note         = {Machine review of arXiv:2508.07141}
}
read the original abstract

Conceptual product design requires designers to explore the design space of visual and functional concepts simultaneously. Sketching has long been adopted to empower concept exploration. However, current sketch-based design tools mostly emphasize visual design using emerging techniques. We present SketchConcept, a design support tool that decomposes design concepts into visual representations and functionality of concepts using sketches and textual descriptions. We propose a function-to-visual mapping workflow that maps the function descriptions generated by a Large Language Model to a component of the concept produced by image Generative Artificial Intelligence(GenAI). The function-to-visual mapping allows our system to leverage multimodal GenAI to decompose, generate, and edit the design concept to satisfy the overall function and behavior. We present multiple use cases enabled by SketchConcept to validate the workflow. Finally, we evaluated the efficacy and usability of our system with a two-session user study.

Figures

Figures reproduced from arXiv: 2508.07141 by the authors.

Figure 1
Figure 1. An overview of the SketchConcept system workflow. The user of SketchConcept wants to design a pink pickup truck. [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Interface layout of the SketchConcept. The user can draw (a-1), view (a-2), and edit (a-3) the concept image on the white [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. An example of concept decomposition. System seg [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Conceptual toy designs of Nerf Gun and Robot Dog using SketchConcept system. (a) input sketch and description; (b) [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]
Figure 5
Figure 5. Figure 5: An example of editing a concept image of home layout in component level. [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: The results of the qualitative feedback on the system [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: Car concept design outcomes from the second session of user study. [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

62 extracted references · 36 canonical work pages

  1. [1]

    Rahul Arora, Rubaiat Habib Kazi, Tovi Grossman, George Fitzmaurice, and Karan Singh. 2018. Symbiosissketch: Combining 2d & 3d sketching for designing detailed 3d objects in situ. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems . 1–15

    work page 2018

  2. [3]

    Seok-Hyung Bae, Ravin Balakrishnan, and Karan Singh. 2008. ILoveSketch: as- natural-as-possible sketching system for creating 3d curve models. InProceedings of the 21st Annual ACM Symposium on User Interface Software and Technology (Monterey, CA, USA) (UIST ’08). Association for Computing Machinery, New York, NY, USA, 151–160. https://doi.org/10.1145/1449...

    arXiv 2008

  3. [4]

    Seok-Hyung Bae, Ravin Balakrishnan, and Karan Singh. 2009. EverybodyLovesS- ketch: 3D sketching for a broader audience. In Proceedings of the 22nd Annual ACM Symposium on User Interface Software and Technology (Victoria, BC, Canada) (UIST ’09). Association for Computing Machinery, New York, NY, USA, 59–68. https://doi.org/10.1145/1622176.1622189

    work page arXiv 2009

  4. [5]

    Mike Baxter. 2018. Product design. CRC press

    work page 2018

  5. [6]

    Nan Cao, Xin Yan, Yang Shi, and Chaoran Chen. 2019. AI-Sketcher : A Deep Generative Model for Producing High-Quality Sketches. Proceedings of the AAAI Conference on Artificial Intelligence 33, 01 (Jul. 2019), 2564–2571. https: //doi.org/10.1609/aaai.v33i01.33012564

    work page doi:10.1609/aaai.v33i01.33012564 2019

  6. [7]

    Liang-Chieh Chen, George Papandreou, Florian Schroff, and Hartwig Adam

    work page

  7. [8]

    Liang-Chieh Chen, Yukun Zhu, George Papandreou, Florian Schroff, and Hartwig Adam. 2018. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. arXiv:1802.02611 [cs.CV]

    arXiv 2018

  8. [10]

    Wei Chen and Mark Fuge. 2021. BézierGAN: Automatic Genera- tion of Smooth Curves from Interpretable Low-Dimensional Parameters. arXiv:1808.08871 [cs.LG]

    work page Pith review arXiv 2021

Show all 62 references
  1. [11]

    Xiang ’Anthony’ Chen, Ye Tao, Guanyun Wang, Runchang Kang, Tovi Grossman, Stelian Coros, and Scott E. Hudson. 2018. Forte: User-Driven Generative Design. In Proceedings of the 2018 CHI Conference on Human Factors in Computing Systems (<conf-loc>, <city>Montreal QC</city>, <cou...

    2018

  2. [12]

    Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. 2009. Im- ageNet: A large-scale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition . 248–255. https://doi.org/10.1109/CVPR. 2009.5206848 SketchConcept: Sketching...

    2009

  3. [13]

    Giulia Di Fede, Davide Rocchesso, Steven P Dow, and Salvatore Andolina. 2022. The Idea Machine: LLM-based Expansion, Rewriting, Combination, and Sugges- tion of Ideas. In Proceedings of the 14th Conference on Creativity and Cognition . 623–627

    2022

  4. [14]

    Tobias Drey, Jan Gugenheimer, Julian Karlbauer, Maximilian Milo, and Enrico Rukzio. 2020. Vrsketchin: Exploring the design space of pen and tablet interaction for 3d sketching in virtual reality. In Proceedings of the 2020 CHI conference on human factors in computing systems . 1–14

    2020

  5. [15]

    Edwards, Brandon Man, and Faez Ahmed

    Kristen M. Edwards, Brandon Man, and Faez Ahmed. 2024. SKETCH2PROTOTYPE: RAPID CONCEPTUAL DESIGN EXPLO- RATION AND PROTOTYPING WITH GENERATIVE AI. Source: http://decode.mit.edu/assets/papers/Sketch2Prototype.pdf. Accessed: April 1, 2024

    2024

  6. [16]

    Dokania, Oliver Wang, Alexei A

    Arnab Ghosh, Richard Zhang, Puneet K. Dokania, Oliver Wang, Alexei A. Efros, Philip H. S. Torr, and Eli Shechtman. 2019. Interactive Sketch & Fill: Multiclass Sketch-to-Image Translation. arXiv:1909.11081 [cs.CV]

    2019 arXiv

  7. [17]

    Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde- Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio

    Ian J. Goodfellow, Jean Pouget-Abadie, Mehdi Mirza, Bing Xu, David Warde- Farley, Sherjil Ozair, Aaron Courville, and Yoshua Bengio. 2014. Generative Adversarial Networks. arXiv:1406.2661 [stat.ML]

    2014 arXiv

  8. [18]

    Ji Han, Hannah Forbes, Feng Shi, Jia Hao, and Dirk Schaefer. 2020. A data-driven approach for creative concept generation and evaluation. In Proceedings of the Design Society: DESIGN Conference , Vol. 1. Cambridge University Press, 167–176

    2020

  9. [20]

    Ji Han, Serhad Sarica, Feng Shi, and Jianxi Luo. 2022. Semantic networks for engineering design: state of the art and future directions. Journal of Mechanical Design 144, 2 (2022), 020802

    2022

  10. [21]

    Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2015. Deep Residual Learning for Image Recognition. arXiv:1512.03385 [cs.CV]

    2015 arXiv

  11. [22]

    Hiroki Kaimoto, Kyzyl Monteiro, Mehrad Faridan, Jiatong Li, Samin Farajian, Yasuaki Kakehi, Ken Nakagaki, and Ryo Suzuki. 2022. Sketched Reality: Sketching Bi-Directional Interactions Between Virtual and Physical Worlds with AR and Actuated Tangible UI. In Proceedings of the 3...

    2022

  12. [23]

    Stahovich

    Levent Burak Kara and Thomas F. Stahovich. 2004. Hierarchical parsing and recognition of hand-sketched diagrams. In Proceedings of the 17th Annual ACM Symposium on User Interface Software and Technology (Santa Fe, NM, USA) (UIST ’04). Association for Computing Machinery, New Y...

    2004

  13. [24]

    Rubaiat Habib Kazi, Tovi Grossman, Hyunmin Cheong, Ali Hashemi, and George Fitzmaurice. 2017. DreamSketch: Early Stage 3D Design Explorations with Sketching and Generative Design. In Proceedings of the 30th Annual ACM Sym- posium on User Interface Software and Technology (Québ...

    2017

  14. [25]

    Sangpil Kim, Hyung-gun Chi, Xiao Hu, Qixing Huang, and Karthik Ramani. 2020. A large-scale annotated mechanical components benchmark for classification and retrieval tasks with deep neural networks. In Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23...

    2020

  15. [26]

    Diederik P Kingma and Max Welling. 2022. Auto-Encoding Variational Bayes. arXiv:1312.6114 [stat.ML]

    2022 arXiv

  16. [27]

    Alexander Kirillov, Ross Girshick, Kaiming He, and Piotr Dollár. 2019. Panoptic Feature Pyramid Networks. arXiv:1901.02446 [cs.CV]

    2019 arXiv

  17. [28]

    Hyung-Kwon Ko, Gwanmo Park, Hyeon Jeon, Jaemin Jo, Juho Kim, and Jinwook Seo. 2023. Large-scale text-to-image generation models for visual artists’ cre- ative works. In Proceedings of the 28th international conference on intelligent user interfaces. 919–933

    2023

  18. [29]

    Landay and B.A

    J.A. Landay and B.A. Myers. 2001. Sketching interfaces: toward more human interface design. Computer 34, 3 (2001), 56–64. https://doi.org/10.1109/2.910894

    2001 doi

  19. [30]

    Bongshin Lee, Rubaiat Habib Kazi, and Greg Smith. 2013. SketchStory: Telling More Engaging Stories with Data through Freeform Sketching. IEEE Transactions on Visualization and Computer Graphics 19, 12 (2013), 2416–2425. https://doi. org/10.1109/TVCG.2013.191

    2013 doi

  20. [31]

    Lawrence Zitnick, and Michael F

    Yong Jae Lee, C. Lawrence Zitnick, and Michael F. Cohen. 2011. ShadowDraw: real-time user guidance for freehand drawing. ACM Trans. Graph. 30, 4, Article 27 (jul 2011), 10 pages. https://doi.org/10.1145/2010324.1964922

    2011

  21. [32]

    Hanchao Li, Pengfei Xiong, Jie An, and Lingxue Wang. 2018. Pyramid Attention Network for Semantic Segmentation. arXiv:1805.10180 [cs.CV]

    2018 arXiv

  22. [33]

    Vivian Liu and Lydia B Chilton. 2022. Design guidelines for prompt engineering text-to-image generative models. In Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems . 1–23

    2022

  23. [34]

    Jianxi Luo, Serhad Sarica, and Kristin L Wood. 2021. Guiding data-driven design ideation by knowledge distance. Knowledge-Based Systems 218 (2021), 106873

    2021

  24. [35]

    Jianxi Luo, Binyang Song, Lucienne Blessing, and Kristin Wood. 2018. Design opportunity conception using the total technology space map. Artificial Intelli- gence for Engineering Design, Analysis and Manufacturing 32 (11 2018), 449–461. https://doi.org/10.1017/S0890060418000094

    2018 doi

  25. [36]

    Kevin Ma, Daniele Grandi, Christopher McComb, and Kosa Goucher-Lambert

  26. [37]

    Yusuke Matsui, Takaaki Shiratori, and Kiyoharu Aizawa. 2017. DrawFromDraw- ings: 2D Drawing Assistance via Stroke Interpolation with a Sketch Database. IEEE Transactions on Visualization and Computer Graphics 23, 7 (2017), 1852–1862. https://doi.org/10.1109/TVCG.2016.2554113

    2017

  27. [38]

    Soumik Mohian and Christoph Csallner. 2022. PSDoodle: searching for app screens via interactive sketching. InProceedings of the 9th IEEE/ACM International Conference on Mobile Software Engineering and Systems (MOBILESoft ’22) . ACM. https://doi.org/10.1145/3524613.3527807

    2022

  28. [39]

    Cecil Piya, Vinayak , Senthil Chandrasegaran, Niklas Elmqvist, and Karthik Ramani. 2017. Co-3Deator: A team-first collaborative 3D design ideation tool. In Proceedings of the 2017 CHI conference on human factors in computing systems . 6581–6592

    2017

  29. [40]

    Devarajan Ramanujan, William Z Bernstein, William Benjamin, Karthik Ramani, Niklas Elmqvist, Devadatta Kulkarni, and Jeffrey Tew. 2015. A framework for visualization-driven eco-conscious design exploration. Journal of Computing and Information Science in Engineering 15, 4 (201...

    2015

  30. [41]

    Olaf Ronneberger, Philipp Fischer, and Thomas Brox. 2015. U-Net: Convolutional Networks for Biomedical Image Segmentation. arXiv:1505.04597 [cs.CV]

    2015 arXiv

  31. [42]

    Sigurdur Gauti Samuelsson and Matthias Book. 2020. Eliciting Sketched Expres- sions of Command Intentions in an IDE. Proc. ACM Hum.-Comput. Interact. 4, ISS, Article 200 (nov 2020), 25 pages. https://doi.org/10.1145/3427328

    2020 doi

  32. [43]

    Serhad Sarica, Ji Han, and Jianxi Luo. 2023. Design representation as semantic networks. Computers in Industry 144 (2023), 103791

    2023

  33. [44]

    Serhad Sarica, Jianxi Luo, and Kristin L. Wood. 2020. TechNet: Technology semantic network based on patent data. Expert Systems with Applications 142 (March 2020), 112995. https://doi.org/10.1016/j.eswa.2019.112995

    2020

  34. [45]

    Ari Seff, Yaniv Ovadia, Wenda Zhou, and Ryan P Adams. 2020. Sketchgraphs: A large-scale dataset for modeling relational geometry in computer-aided design. arXiv preprint arXiv:2007.08506 (2020)

    2020 arXiv

  35. [46]

    Ari Seff, Wenda Zhou, Nick Richardson, and Ryan P Adams. 2021. Vitruvion: A generative model of parametric cad sketches. arXiv preprint arXiv:2109.14124 (2021)

    2021 arXiv

  36. [47]

    Feng Shi, Liuqing Chen, Ji Han, and Peter Childs. 2017. A Data-Driven Text Mining and Semantic Network Analysis for Design Information Retrieval. Jour- nal of Mechanical Design 139, 11 (10 2017), 111402. https://doi.org/10.1115/1. 4037649 arXiv:https://asmedigitalcollection.as...

    2017 doi

  37. [48]

    Yang Shi, Nan Cao, Xiaojuan Ma, Siji Chen, and Pei Liu. 2020. EmoG: Supporting the Sketching of Emotional Expressions for Storyboarding. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems (<conf-loc>, <city>Honolulu</city>, <state>HI</state>, <coun...

    2020

  38. [49]

    Shuang Song and Alice M Agogino. 2004. Insights on designers’ sketching activi- ties in new product design teams. In International Design Engineering Technical Conferences and Computers and Information in Engineering Conference , Vol. 46962. 351–360

    2004

  39. [50]

    Ryo Suzuki, Rubaiat Habib Kazi, Li-yi Wei, Stephen DiVerdi, Wilmot Li, and Daniel Leithinger. 2020. RealitySketch: Embedding Responsive Graphics and Visualizations in AR through Dynamic Sketching. In Proceedings of the 33rd Annual ACM Symposium on User Interface Software and T...

    2020

  40. [51]

    David G Ullman, Stephen Wood, and David Craig. 1990. The importance of drawing in the mechanical design process. Computers & graphics 14, 2 (1990), 263–274

    1990

  41. [52]

    Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems 30 (2017)

    2017

  42. [53]

    Mathias Peter Verheijden and Mathias Funk. 2023. Collaborative Diffusion: Boosting Designerly Co-Creation with Generative AI. In Extended Abstracts of the 2023 CHI Conference on Human Factors in Computing Systems (<conf-loc>, <city>Hamburg</city>, <country>Germany</country>, <...

    2023

  43. [54]

    Fang Wang, Le Kang, and Yi Li. 2015. Sketch-based 3D Shape Retrieval using Convolutional Neural Networks. arXiv:1504.03504 [cs.CV]

    2015 arXiv

  44. [55]

    Chenfei Wu, Shengming Yin, Weizhen Qi, Xiaodong Wang, Zecheng Tang, and Nan Duan. 2023. Visual chatgpt: Talking, drawing and editing with visual foun- dation models. arXiv preprint arXiv:2303.04671 (2023). PrePrint for Conference acronym ’XX, June 03–05, 2018, Woodstock, NY Ru...

    2023 arXiv

  45. [56]

    Haijun Xia, Nathalie Henry Riche, Fanny Chevalier, Bruno De Araujo, and Daniel Wigdor. 2018. DataInk: Direct and Creative Data-Oriented Drawing. In Pro- ceedings of the 2018 CHI Conference on Human Factors in Computing Systems (<conf-loc>, <city>Montreal QC</city>, <country>Ca...

    2018

  46. [57]

    Mitra, Daniel Cohen-Or, Wenyong Gong, Qi Su, and Baoquan Chen

    Xiaohua Xie, Kai Xu, Niloy J. Mitra, Daniel Cohen-Or, Wenyong Gong, Qi Su, and Baoquan Chen. 2013. Sketch-to-Design: Context-Based Part Assembly. Computer Graphics Forum 32, 8 (Oct. 2013), 233–245. https://doi.org/10.1111/cgf.12200

    2013 doi

  47. [58]

    Emre Yilmaz and Brian German. [n. d.]. Conditional Generative Adversarial Network Framework for Airfoil Inverse Design. https://doi.org/10.2514/6.2020-3185 arXiv:https://arc.aiaa.org/doi/pdf/10.2514/6.2020-3185

    2020 doi

  48. [59]

    Wentai Zhang, Zhangsihao Yang, Haoliang Jiang, Suyash Nigam, Soji Yamakawa, Tomotake Furuhata, Kenji Shimada, and Levent Burak Kara. 2019. 3D Shape Syn- thesis for Conceptual Design and Optimization Using Variational Autoencoders. arXiv:1904.07964 [cs.LG]

    2019 arXiv

  49. [60]

    Zongwei Zhou, Md Mahfuzur Rahman Siddiquee, Nima Tajbakhsh, and Jian- ming Liang. 2018. UNet++: A Nested U-Net Architecture for Medical Image Segmentation. arXiv:1807.10165 [cs.CV]

    2018 arXiv

  50. [61]

    Qihao Zhu and Jianxi Luo. 2023. Generative transformers for design concept generation. Journal of Computing and Information Science in Engineering 23, 4 (2023), 041003

    2023

  51. [62]

    Qihao Zhu, Xinyu Zhang, and Jianxi Luo. 2023. Biologically Inspired Design Concept Generation Using Generative Pre-Trained Transformers. Journal of Mechanical Design 145, 4 (2023), 041409

    2023

  52. [63]

    Zhengzhe Zhu, Ziyi Liu, Tianyi Wang, Youyou Zhang, Xun Qian, Pashin Farsak Raja, Ana Villanueva, and Karthik Ramani. 2022. MechARspace: An authoring system enabling bidirectional binding of augmented reality with toys in real-time. In Proceedings of the 35th Annual ACM Symposi...

    2022

  53. [2017]

    arXiv:1706.05587 [cs.CV]

    Rethinking Atrous Convolution for Semantic Image Segmentation. arXiv:1706.05587 [cs.CV]

    arXiv

  54. [2023]

    arXiv preprint arXiv:2306.01779 (2023)

    Conceptual Design Generation Using Large Language Models. arXiv preprint arXiv:2306.01779 (2023)

    2023 arXiv

Pith tools

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