Food Label Evidence

Measurement

Why you cannot estimate a portion

Portion estimation error is large enough to swamp most of what people try to change about their diet. This is the problem food-scanning apps exist to solve.

Updated 3 min read 17 citations Evidence strength 4/5

The size of the problem

Systematic review of portion size estimation [3] and of recall errors [2] finds errors that routinely reach tens of percent. Crucially the error is biased, not random: people under-report, and under-report more when intake is higher and when the food is one they feel judged for.

Validation against doubly labelled water — the reference method, covered on our metabolism site — quantifies this directly. It is why a food diary showing 1,600 kcal alongside no weight change is usually a measurement problem rather than a metabolic one.

Portion estimation error

Why it is hard

  • Volume perception is non-linear. Doubling a bowl's diameter more than doubles its contents, and people judge by width.
  • Cooking oil is invisible. A tablespoon is over 100 kcal and appears in no memory of a meal.
  • Packaging defines portions. A "portion" is whatever the container held.
  • Reference objects fail. "Palm-sized" varies with the palm.
  • Memory reconstructs. People recall what they usually eat, not what they ate.

Training helps, and a systematic review of training effects [4] shows the improvement decays without reinforcement — the usual finding for a perceptual skill.

Can technology fix it?

This is where image-based dietary assessment comes in: photograph the meal, let a model identify foods and estimate volume. Validation studies against reference methods exist, and augmented reality has been reviewed as an estimation aid [1].

Honest summary: image-based methods are imperfect and better than recall. They remove the memory step, which is the largest error source, and they inherit new errors around depth estimation and hidden ingredients. Against a food diary written from memory at bedtime, that is a clear improvement. Against a kitchen scale, it is not.

What this means for tracking

Use tracked calories as a relative signal, not an absolute one. If your logged intake is consistent week to week and your weight is not moving, the useful move is to change something and watch the trend — not to trust the integer. Consistency of method matters far more than accuracy of method, because a consistent bias cancels when you compare against yourself.

Common questions

How wrong is my food diary?
Systematically under, often by tens of percent, and more so at higher intakes [2][3].
Are photo-based apps accurate?
Better than recall, worse than weighing. They remove the memory error and add estimation error.
Should I weigh everything?
For a few weeks, it is genuinely educational. Indefinitely, most people stop — and a method you abandon has an accuracy of zero.
Why does my app say I ate 1,500 kcal and I am not losing?
Almost always under-reporting rather than metabolism. See our metabolism site.

References

Every citation below links to the original peer-reviewed record on PubMed or via DOI. Nothing here is a substitute for medical advice.

  1. Application of Augmented Reality Technology as a Dietary Monitoring and Control Measure Among Adults: A Systematic Review Gonzalez GV, Mao B, Wang R, et al. · Nutrients · 2025 · Systematic review DOIPubMed 41470837Full text
  2. A systematic review of recall errors associated with portion size estimation aids in children de Vlieger NM, Weltert M, Molenaar A, et al. · Appetite · 2020 · Systematic review DOIPubMed 31751634
  3. Validity of image-based dietary assessment methods: A systematic review and meta-analysis Ho DKN, Tseng SH, Wu MC, et al. · Clinical nutrition (Edinburgh, Scotland) · 2020 · Meta-analysis DOIPubMed 32839035
  4. Portion size estimation in dietary assessment: a systematic review of existing tools, their strengths and limitations Amoutzopoulos B, Page P, Roberts C, et al. · Nutrition reviews · 2020 · Systematic review DOIPubMed 31999347
  5. The Role of Various Forms of Training on Improved Accuracy of Food-Portion Estimation Skills: A Systematic Review of the Literature Hooper A, McMahon A, Probst Y · Advances in nutrition (Bethesda, Md.) · 2019 · Systematic review DOIPubMed 30629097Full text
  6. Dietary assessment in minority ethnic groups: a systematic review of instruments for portion-size estimation in the United Kingdom Almiron-Roig E, Aitken A, Galloway C, et al. · Nutrition reviews · 2017 · Systematic review DOIPubMed 28340101Full text
  7. Immersive Virtual Reality Dietitian Improves Portion Control Self-Efficacy and Portion Size Estimation Accuracy Braga BC, Long J, Maksi S, et al. · Journal of nutrition education and behavior · 2025 · Randomised controlled trial DOIPubMed 40019425
  8. An automated image-based dietary assessment application: a pilot study Lee L, Bishop R, Stanley J · Journal of nutritional science · 2025 · Randomised controlled trial DOIPubMed 41321681Full text
  9. Validity and Usability of a Smartphone Image-Based Dietary Assessment App Compared to 3-Day Food Diaries in Assessing Dietary Intake Among Canadian Adults: Randomized Controlled Trial Ji Y, Plourde H, Bouzo V, et al. · JMIR mHealth and uHealth · 2020 · Randomised controlled trial DOIPubMed 32902389Full text
  10. The international food unit: a new measurement aid that can improve portion size estimation Bucher T, Weltert M, Rollo ME, et al. · The international journal of behavioral nutrition and physical activity · 2017 · Randomised controlled trial DOIPubMed 28899402Full text
  11. Portion-size estimation training in second- and third-grade American Indian children Weber JL, Cunningham-Sabo L, Skipper B, et al. · The American journal of clinical nutrition · 1999 · Randomised controlled trial DOIPubMed 10195603
  12. New mobile methods for dietary assessment: review of image-assisted and image-based dietary assessment methods Boushey CJ, Spoden M, Zhu FM, et al. · The Proceedings of the Nutrition Society · 2017 · Review DOIPubMed 27938425
  13. Merging dietary assessment with the adolescent lifestyle Schap TE, Zhu F, Delp EJ, et al. · Journal of human nutrition and dietetics : the official journal of the British Dietetic Association · 2014 · Review DOIPubMed 23489518Full text
  14. Development and validation of a smartphone image-based app for dietary intake assessment among Palestinian undergraduates Hattab S, Badrasawi M, Anabtawi O, et al. · Scientific reports · 2022 · Journal article DOIPubMed 36104377Full text
  15. Development and evaluation of image-series for portion size estimation in dietary assessment among adults Salvesen L, Engeset D, Øverby NC, et al. · Journal of nutritional science · 2021 · Journal article DOIPubMed 33889386Full text
  16. The accuracy of portion size estimation using food images and textual descriptions of portion sizes: an evaluation study Lucassen DA, Willemsen RF, Geelen A, et al. · Journal of human nutrition and dietetics : the official journal of the British Dietetic Association · 2021 · Journal article DOIPubMed 33761165Full text
  17. Validation of a Smartphone Image-Based Dietary Assessment Method for Pregnant Women Ashman AM, Collins CE, Brown LJ, et al. · Nutrients · 2017 · Journal article DOIPubMed 28106758Full text