Food Image Segmentation with LLM-Derived Ingredient Labels and Multimodal Fusion

• URL: http://arxiv.org/abs/2607.25820v1
• Date: Tue, 28 Jul 2026 15:02:46 GMT
• Title: Food Image Segmentation with LLM-Derived Ingredient Labels and Multimodal Fusion
• Abstract summary: We propose two plug-and-play multimodal modules that enhance the segmentation performance.<n>LIM-F and LIM-Q inject semantic ingredient information into the visual analysis pipeline.<n>The proposed approach offers a practical and scalable solution for fine-grained food understanding.
• Score: 3.8615905456206256
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Food image segmentation plays a vital role in health-related applications such as nutrition tracking and personalized health monitoring. However, existing models often underperform on visually similar ingredients and rare food categories. To address this issue, we propose two plug-and-play multimodal modules that enhance the segmentation performance by leveraging ingredient labels inferred from food images using large language models (LLMs). The first module, called LIM-F (Language Injection Module for Features), is designed to pair with any image encoder that produces multi-layer outputs (e.g., Swin Transformer), while the second module, LIM-Q (Language Injection Module for Queries), targets Mask2Former-style Transformer-based decoders. Both modules enable training without the need for pre-aligning images with text by directly injecting semantic ingredient information into the visual analysis pipeline. On the FoodSeg103 benchmark, the proposed method achieves state-of-the-art performance. Specifically, integrating LIM-Q into the Mask2Former decoder with a Swin-L image encoder yields a mean Intersection over Union (mIoU) of 55.0. LIM-F also demonstrates strong generalization and competitive performance, reaching an mIoU of 54.4 under the same model (Swin-L+Mask2Former). Furthermore, its applicability extends beyond Transformer-based decoders, as evidenced by an improvement from 47.7 to 49.8 mIoU when integrated into a CNN-based architecture. Notably, the improved segmentation accuracy is achieved with only a moderate (at most 3.8 GB) increase in the GPU memory consumption during training. Thus, the proposed approach offers a practical and scalable solution for fine-grained food understanding.

Related papers

• Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation [66.53544128707817]
  Cheers is a unified multimodal model that decouples patch-level details from semantic representations.<n> Cheers matches or surpasses advanced UMMs in both visual understanding and generation.
  arXiv  Detail & Related papers  (2026-03-13T08:55:27Z)
• Weakly Supervised Food Image Segmentation using Vision Transformers and Segment Anything Model [45.641643140398735]
  We propose a weakly supervised semantic segmentation approach for food images.<n>We use the Segment Anything Model (SAM) along with the attention mechanisms of Vision Transformers (ViTs)<n>We envision the proposed approach as a tool to accelerate food image annotation tasks or as an integrated component in food and nutrition tracking applications.
  arXiv  Detail & Related papers  (2025-09-23T14:01:51Z)
• ContextFormer: Redefining Efficiency in Semantic Segmentation [48.81126061219231]
  Convolutional methods, although capturing local dependencies well, struggle with long-range relationships.<n>Vision Transformers (ViTs) excel in global context capture but are hindered by high computational demands.<n>We propose ContextFormer, a hybrid framework leveraging the strengths of CNNs and ViTs in the bottleneck to balance efficiency, accuracy, and robustness for real-time semantic segmentation.
  arXiv  Detail & Related papers  (2025-01-31T16:11:04Z)
• MSDNet: Multi-Scale Decoder for Few-Shot Semantic Segmentation via Transformer-Guided Prototyping [1.1852406625172218]
  Few-shot Semantic addresses the challenge of segmenting objects in query images with only a handful of annotated examples.<n>We propose a new Few-shot Semantic framework based on the Transformer architecture.
  arXiv  Detail & Related papers  (2024-09-17T16:14:03Z)
• MagDiff: Multi-Alignment Diffusion for High-Fidelity Video Generation and Editing [90.06041718086317]
  We propose a unified Multi-alignment Diffusion, dubbed as MagDiff, for both tasks of high-fidelity video generation and editing. The proposed MagDiff introduces three types of alignments, including subject-driven alignment, adaptive prompts alignment, and high-fidelity alignment.
  arXiv  Detail & Related papers  (2023-11-29T03:36:07Z)
• BatchFormerV2: Exploring Sample Relationships for Dense Representation Learning [88.82371069668147]
  BatchFormerV2 is a more general batch Transformer module, which enables exploring sample relationships for dense representation learning. BatchFormerV2 consistently improves current DETR-based detection methods by over 1.3%.
  arXiv  Detail & Related papers  (2022-04-04T05:53:42Z)
• VLMo: Unified Vision-Language Pre-Training with Mixture-of-Modality-Experts [46.55920956687346]
  We present a unified Vision-Language pretrained Model (VLMo) that jointly learns a dual encoder and a fusion encoder with a modular Transformer network. Because of the modeling flexibility of MoME, pretrained VLMo can be fine-tuned as a fusion encoder for vision-language classification tasks. We propose a stagewise pre-training strategy, which effectively leverages large-scale image-only and text-only data besides image-text pairs.
  arXiv  Detail & Related papers  (2021-11-03T17:20:36Z)
• Boosting Few-shot Semantic Segmentation with Transformers [81.43459055197435]
  TRansformer-based Few-shot Semantic segmentation method (TRFS) Our model consists of two modules: Global Enhancement Module (GEM) and Local Enhancement Module (LEM)
  arXiv  Detail & Related papers  (2021-08-04T20:09:21Z)
• Segmenter: Transformer for Semantic Segmentation [79.9887988699159]
  We introduce Segmenter, a transformer model for semantic segmentation. We build on the recent Vision Transformer (ViT) and extend it to semantic segmentation. It outperforms the state of the art on the challenging ADE20K dataset and performs on-par on Pascal Context and Cityscapes.
  arXiv  Detail & Related papers  (2021-05-12T13:01:44Z)
• Prototype Mixture Models for Few-shot Semantic Segmentation [50.866870384596446]
  Few-shot segmentation is challenging because objects within the support and query images could significantly differ in appearance and pose. We propose prototype mixture models (PMMs), which correlate diverse image regions with multiple prototypes to enforce the prototype-based semantic representation. PMMs improve 5-shot segmentation performance on MS-COCO by up to 5.82% with only a moderate cost for model size and inference speed.
  arXiv  Detail & Related papers  (2020-08-10T04:33:17Z)

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.