Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Diffusion Models for Generative Outfit Recommendation (2402.17279v3)

Published 27 Feb 2024 in cs.IR

Abstract: Outfit Recommendation (OR) in the fashion domain has evolved through two stages: Pre-defined Outfit Recommendation and Personalized Outfit Composition. However, both stages are constrained by existing fashion products, limiting their effectiveness in addressing users' diverse fashion needs. Recently, the advent of AI-generated content provides the opportunity for OR to transcend these limitations, showcasing the potential for personalized outfit generation and recommendation. To this end, we introduce a novel task called Generative Outfit Recommendation (GOR), aiming to generate a set of fashion images and compose them into a visually compatible outfit tailored to specific users. The key objectives of GOR lie in the high fidelity, compatibility, and personalization of generated outfits. To achieve these, we propose a generative outfit recommender model named DiFashion, which empowers exceptional diffusion models to accomplish the parallel generation of multiple fashion images. To ensure three objectives, we design three kinds of conditions to guide the parallel generation process and adopt Classifier-Free-Guidance to enhance the alignment between the generated images and conditions. We apply DiFashion on both personalized Fill-In-The-Blank and GOR tasks and conduct extensive experiments on iFashion and Polyvore-U datasets. The quantitative and human-involved qualitative evaluation demonstrate the superiority of DiFashion over competitive baselines.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (49)
  1. Manal A. Alshehri and Xiangliang Zhang. 2022. Generative Adversarial Zero-Shot Learning for Cold-Start News Recommendation. In CIKM. ACM, 26–36.
  2. Personalized Bundle List Recommendation. In WWW. ACM, 60–71.
  3. Multimodal Garment Designer: Human-Centric Latent Diffusion Models for Fashion Image Editing. In arXiv:2304.02051.
  4. Fashion Outfit Generation for E-Commerce. In ECML/PKDD. Springer, 339–354.
  5. InstructPix2Pix: Learning to Follow Image Editing Instructions. In arXiv:2211.09800.
  6. Language models are few-shot learners. NeurIPS 33 (2020), 1877–1901.
  7. Image Reference-Guided Fashion Design With Structure-Aware Transfer by Diffusion Models. In CVPRW. IEEE, 3524–3528.
  8. Outfit Generation and Recommendation—An Experimental Study. In FashionxRecsys. ACM, 117–137.
  9. PixArt-α𝛼\alphaitalic_α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis. In arXiv:2310.00426.
  10. POG: Personalized Outfit Generation for Fashion Recommendation at Alibaba IFashion. In KDD. ACM, 2662–2670.
  11. Dressing as a Whole: Outfit Compatibility Learning Based on Node-wise Graph Neural Networks. In WWW. ACM, 307–317.
  12. Prafulla Dhariwal and Alexander Nichol. 2021. Diffusion models beat gans on image synthesis. NeurIPS 34 (2021), 8780–8794.
  13. Computational Technologies for Fashion Recommendation: A Survey. ACM Comput. Surv. 56, 5 (2023).
  14. Personalized Fashion Outfit Generation with User Coordination Preference Learning. IPM 60, 5 (2023), 103434.
  15. Personalized Capsule Wardrobe Creation with Garment and User Modeling. In MM. ACM, 302–310.
  16. AI Assisted Fashion Design: A Review. IEEE Access (2023).
  17. Learning Fashion Compatibility with Bidirectional LSTMs. In MM. ACM, 1078–1086.
  18. Lightgcn: Simplifying and powering graph convolution network for recommendation. In SIGIR. ACM, 639–648.
  19. Denoising diffusion probabilistic models. NeurIPS 33 (2020), 6840–6851.
  20. Jonathan Ho and Tim Salimans. 2022. Classifier-free diffusion guidance. In arXiv:2207.12598.
  21. Collaborative Fashion Recommendation: A Functional Tensor Factorization Approach. In MM. ACM, 129–138.
  22. Text to image generation of fashion clothing. In INDIACom. IEEE, 355–358.
  23. Leveraging off-the-shelf diffusion model for multi-attribute fashion image manipulation. In WACV. IEEE, 848–857.
  24. Coherent and controllable outfit generation. In arXiv:1906.07273.
  25. Hierarchical Fashion Graph Network for Personalized Outfit Recommendation. In SIGIR. ACM, 159–168.
  26. Variational Autoencoders for Collaborative Filtering. In WWW. ACM, 689–698.
  27. OutfitNet: Fashion Outfit Recommendation with Attention-Based Multiple Instance Learning. In WWW. ACM, 77–87.
  28. ONCE: Boosting Content-based Recommendation with Both Open- and Closed-source Large Language Models.
  29. Personalized Outfit Recommendation with Learnable Anchors. In CVPR. IEEE, 12717–12726.
  30. Learning Binary Code for Personalized Fashion Recommendation. In CVPR. IEEE, 10554–10562.
  31. Calvin Luo. 2022. Understanding diffusion models: A unified perspective. In arXiv:2208.11970.
  32. OutfitGAN: Learning Compatible Items for Generative Fashion Outfits. In CVPRW. IEEE, 2272–2276.
  33. Training language models to follow instructions with human feedback. NeurIPS 35 (2022), 27730–27744.
  34. William Peebles and Saining Xie. 2023. Scalable diffusion models with transformers. In ICCV. IEEE, 4195–4205.
  35. BPR: Bayesian personalized ranking from implicit feedback. In UAI. AUAI Press, 452–461.
  36. High-resolution image synthesis with latent diffusion models. In CVPR. 10684–10695.
  37. Outfittransformer: Learning outfit representations for fashion recommendation. In WACV. IEEE, 3601–3609.
  38. Development of an intelligent data-driven system to recommend personalized fashion design solutions. Sensors 21, 12 (2021), 4239.
  39. SGDiff: A Style Guided Diffusion Model for Fashion Synthesis. In MM. ACM, 8433–8442.
  40. Rethinking the inception architecture for computer vision. In CVPR. IEEE, 2818–2826.
  41. Generative recommendation: Towards next-generation recommender paradigm. In arXiv:2304.03516.
  42. Diffusion Recommender Model. In SIGIR. ACM, 832–841.
  43. FashionDiff: A Controllable Diffusion Model Using Pairwise Fashion Elements for Intelligent Design. In MM. ACM, 1401–1411.
  44. A New Creative Generation Pipeline for Click-Through Rate with Stable Diffusion Model. In arXiv:2401.10934.
  45. Generate What You Prefer: Reshaping Sequential Recommendation via Guided Diffusion. In arXiv:2310.20453.
  46. Personalized Fashion Design. In ICCV. IEEE, 9046–9055.
  47. A3superscript𝐴3A^{3}italic_A start_POSTSUPERSCRIPT 3 end_POSTSUPERSCRIPT-FKG: Attentive Attribute-Aware Fashion Knowledge Graph for Outfit Preference Prediction. IEEE Multimedia 24 (2021), 819–831.
  48. Adding conditional control to text-to-image diffusion models. In ICCV. IEEE, 3836–3847.
  49. The unreasonable effectiveness of deep features as a perceptual metric. In CVPR. IEEE, 586–595.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (6)
  1. Yiyan Xu (12 papers)
  2. Wenjie Wang (150 papers)
  3. Fuli Feng (143 papers)
  4. Yunshan Ma (43 papers)
  5. Jizhi Zhang (24 papers)
  6. Xiangnan He (200 papers)
Citations (8)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now