Training-Free Text-to-Image Compositional Food Generation via Prompt Grafting

• URL: http://arxiv.org/abs/2601.17666v1
• Date: Sun, 25 Jan 2026 03:07:17 GMT
• Title: Training-Free Text-to-Image Compositional Food Generation via Prompt Grafting
• Authors: Xinyue Pan, Yuhao Chen, Fengqing Zhu,
• Abstract summary: Real-world meal images often contain multiple food items.<n>Modern text-to-image diffusion models struggle to generate accurate multi-food images due to object entanglement.<n>We introduce Prompt Grafting, a training-free framework that combines explicit spatial cues in text with implicit layout guidance during sampling.
• Score: 13.309829477759527
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Real-world meal images often contain multiple food items, making reliable compositional food image generation important for applications such as image-based dietary assessment, where multi-food data augmentation is needed, and recipe visualization. However, modern text-to-image diffusion models struggle to generate accurate multi-food images due to object entanglement, where adjacent foods (e.g., rice and soup) fuse together because many foods do not have clear boundaries. To address this challenge, we introduce Prompt Grafting (PG), a training-free framework that combines explicit spatial cues in text with implicit layout guidance during sampling. PG runs a two-stage process where a layout prompt first establishes distinct regions and the target prompt is grafted once layout formation stabilizes. The framework enables food entanglement control: users can specify which food items should remain separated or be intentionally mixed by editing the arrangement of layouts. Across two food datasets, our method significantly improves the presence of target objects and provides qualitative evidence of controllable separation.

Related papers

• CookAnything: A Framework for Flexible and Consistent Multi-Step Recipe Image Generation [34.977083209936815]
  CookAnything is a framework that generates coherent, semantically distinct image sequences from cooking instructions of arbitrary length.<n>It supports scalable, high-quality visual synthesis of complex multi-step instructions and holds significant potential for broad applications in instructional media, and procedural content creation.
  arXiv  Detail & Related papers  (2025-12-03T08:01:48Z)
• LLMs-based Augmentation for Domain Adaptation in Long-tailed Food Datasets [54.527878056610156]
  We present a framework empowered with large language models (LLMs) to address these challenges in food recognition.<n>We first leverage LLMs to parse food images to generate food titles and ingredients.<n>Then, we project the generated texts and food images from different domains to a shared embedding space to maximize the pair similarities.
  arXiv  Detail & Related papers  (2025-11-20T04:38:56Z)
• CookingDiffusion: Cooking Procedural Image Generation with Stable Diffusion [58.92430755180394]
  We present textbfCookingDiffusion, a novel approach to generate photo-realistic images of cooking steps.<n>These prompts encompass text prompts, image prompts, and multi-modal prompts, ensuring the consistent generation of cooking procedural images.<n>Our experimental results demonstrate that our model excels at generating high-quality cooking procedural images.
  arXiv  Detail & Related papers  (2025-01-15T06:58:53Z)
• Foodfusion: A Novel Approach for Food Image Composition via Diffusion Models [48.821150379374714]
  We introduce a large-scale, high-quality food image composite dataset, FC22k, which comprises 22,000 foreground, background, and ground truth ternary image pairs. We propose a novel food image composition method, Foodfusion, which incorporates a Fusion Module for processing and integrating foreground and background information.
  arXiv  Detail & Related papers  (2024-08-26T09:32:16Z)
• OVFoodSeg: Elevating Open-Vocabulary Food Image Segmentation via Image-Informed Textual Representation [43.65207396061584]
  OVFoodSeg is a framework that enhances text embeddings with visual context. The training process of OVFoodSeg is divided into two stages: the pre-training of FoodLearner and the subsequent learning phase for segmentation. By addressing the deficiencies of previous models, OVFoodSeg demonstrates a significant improvement, achieving a 4.9% increase in mean Intersection over Union (mIoU) on the FoodSeg103 dataset.
  arXiv  Detail & Related papers  (2024-04-01T18:26:29Z)
• Transferring Knowledge for Food Image Segmentation using Transformers and Convolutions [65.50975507723827]
  Food image segmentation is an important task that has ubiquitous applications, such as estimating the nutritional value of a plate of food. One challenge is that food items can overlap and mix, making them difficult to distinguish. Two models are trained and compared, one based on convolutional neural networks and the other on Bidirectional representation for Image Transformers (BEiT) The BEiT model outperforms the previous state-of-the-art model by achieving a mean intersection over union of 49.4 on FoodSeg103.
  arXiv  Detail & Related papers  (2023-06-15T15:38:10Z)
• A Large-Scale Benchmark for Food Image Segmentation [62.28029856051079]
  We build a new food image dataset FoodSeg103 (and its extension FoodSeg154) containing 9,490 images. We annotate these images with 154 ingredient classes and each image has an average of 6 ingredient labels and pixel-wise masks. We propose a multi-modality pre-training approach called ReLeM that explicitly equips a segmentation model with rich and semantic food knowledge.
  arXiv  Detail & Related papers  (2021-05-12T03:00:07Z)
• An End-to-End Food Image Analysis System [8.622335099019214]
  We propose an image-based food analysis framework that integrates food localization, classification and portion size estimation. Our proposed framework is end-to-end, i.e., the input can be an arbitrary food image containing multiple food items. Our framework is evaluated on a real life food image dataset collected from a nutrition feeding study.
  arXiv  Detail & Related papers  (2021-02-01T05:36:20Z)
• Cross-Modal Food Retrieval: Learning a Joint Embedding of Food Images and Recipes with Semantic Consistency and Attention Mechanism [70.85894675131624]
  We learn an embedding of images and recipes in a common feature space, such that the corresponding image-recipe embeddings lie close to one another. We propose Semantic-Consistent and Attention-based Networks (SCAN), which regularize the embeddings of the two modalities through aligning output semantic probabilities. We show that we can outperform several state-of-the-art cross-modal retrieval strategies for food images and cooking recipes by a significant margin.
  arXiv  Detail & Related papers  (2020-03-09T07:41:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.