Research map / Vision-language and multimodal / Human and embodied modeling
Nutrition5k: Towards Automatic Nutritional Understanding of Generic Food
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use per-gram nutrient prediction and scale by ground-truth mass instead of direct absolute regression | Nutrition5k RGB nutrition prediction · Nutri-Test | Calorie MAE: 70.6 / 26.1% -> 24.1 / 9.5% | Helped |
| Add depth as a fourth input channel (ablation) | 2D Direct Prediction · Nutri-Test | Calorie MAE: 70.6 / 26.1% -> 47.6 / 18.8% | Helped |
| Add depth-derived volume scalar to image-based mass regression (ablation) | Mass prediction · Nutri-Test RGB-D subset | Total Mass MAE: 38.1g / 29.5% -> 29.4g / 13.7% | Helped |
| Use image+volume mass prediction with per-gram nutrient predictions | End-to-end nutrition prediction · Nutri-Test | Calorie MAE: 70.6 / 26.1% -> 41.3 / 16.5% | Helped |
| Estimate portion size visually rather than using the model | Nutritionists estimating dish mass · 10 Nutri-Test images | Mass percent error: 41% | Hurt |
| Estimate portion size visually rather than using the model | Non-nutritionist annotators · 10 Nutri-Test images | Mass percent error: 53% | Hurt |
Closest papers by meaning
- NutriBench: A Dataset for Evaluating Large Language Models in Nutrition Estimation from Meal Descriptions (ICLR 2025 · similarity 0.86)
- OVFoodSeg: Elevating Open-Vocabulary Food Image Segmentation via Image-Informed Textual Representation (CVPR 2024 · similarity 0.83)
- Pic2Prep: A Multimodal Conversational Agent for Cooking Assistance (AAAI 2025 · similarity 0.82)
- Leveraging Computer Vision and Visual LLMs for Cost-Effective and Consistent Street Food Safety Assessment in Kolkata India (AAAI 2025 · similarity 0.81)
- HD-EPIC: A Highly-Detailed Egocentric Video Dataset (CVPR 2025 · similarity 0.80)
- ViTamin: Designing Scalable Vision Models in the Vision-Language Era (CVPR 2024 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.