Grounding Free-Form Instructions for Fashion Complementary Image Generation
2026-08-24 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors study how to create clothing images that match a given item based on user instructions spoken naturally, instead of using fixed phrases like "a photo of a skirt." They improved existing datasets by adding example instructions with different detail levels and checked these with human reviewers. They designed a new model called StyleFlow that looks at both the original clothing image and the user’s words to make new matching clothing. Their tests show StyleFlow makes better-matching clothes with less complexity and faster running time than other methods. Overall, their work helps computers understand natural language and images together in fashion design.
complementary image generationmultimodal groundingnatural language instructionsfashion datasetsStyleFlowRectified Flow Matchingmultimodal transformerimage quality metricscatalog-alignmenthuman evaluation
Authors
Matteo Attimonelli, Claudio Pomo, Alessandro De Bellis, Danilo Danese, Dietmar Jannach, Tommaso Di Noia
Abstract
Fashion complementary image generation (CIG) aims to create garments that stylistically match a seed item based on user intent, making it a natural multimodal grounding problem where models must interpret language in visual context. Existing CIG benchmarks rely on rigid template prompts (e.g., "a photo of a skirt"), failing to reflect natural user queries and obscuring model behavior across levels of linguistic specificity. We introduce fashion complementary image generation with free-form instructions, a multimodal language-grounding setting where a model generates a compatible garment from a seed image and a natural-language instruction. To this end, we enrich three CIG benchmarks with low-, medium-, and high-specificity instructions generated by a vision-language model and validated by human annotators. We instantiate the task with StyleFlow, a Rectified Flow Matching model that jointly conditions on the seed image and instruction within a single multimodal transformer. Across image quality metrics, catalog-alignment analysis, ablations, and human evaluation, StyleFlow consistently produces instruction-aligned and stylistically coherent garments while reducing architectural complexity and inference cost relative to auxiliary-module approaches.