Rankings › Foodvisor

Rank #6 of 7

Foodvisor Review

A Paris-based AI food tracker with excellent European cuisine coverage — but poor portion accuracy and no chat logging limit its appeal outside its home market.

Rank #6 of 7 – 2026 Benchmark

Foodvisor

European-focused AI calorie tracker with strong French and Mediterranean food databases. Portion estimation relies on 2D scaling only — the main driver of its high MAPE.

6.8
/ 10 Overall

Performance Breakdown

Recognition Accuracy
5.9/10
Portion Estimation
4.0/10
Processing Speed
6.5/10
Category Coverage
8.0/10
Learning & Adaptation
5.0/10
⬇ App Store ⬇ Google Play vs. Welling →

App Highlights

ID Rate
58.9%
Portion MAPE
±31%
Median Speed
7.3s
Food Categories
2,600+
European Cuisine
Excellent
AI Coaching
No
Depth Estimation
No (2D only)
Platforms
iOS + Android
Strengths

Where Foodvisor Excels

🥐

European Cuisine Depth

Best European food database in our test — French, Italian, Spanish, and Northern European dishes identified with higher accuracy than any other tested app.

🗃️

Large Category Database

2,600+ food categories — second only to MyFitnessPal's count, with particular depth in Western European regional specialties.

👩‍⚕️

Registered Dietitian Option

Premium plan includes optional access to a registered dietitian for personalised meal plans — a unique offering among photo-based trackers.

Verdict

Pros & Cons

✓ Pros

  • ✓ Best European & French cuisine coverage
  • ✓ Large 2,600+ food database
  • ✓ Optional access to registered dietitian

✗ Cons

  • ✗ 58.9% overall ID rate — lowest bar 2
  • ✗ ±31% portion MAPE — 2D scaling only
  • ✗ Weak outside Europe — poor global coverage
  • ✗ No AI coaching or chat logging
FAQ

Foodvisor FAQ

Is Foodvisor good for European users?▼
Better than most competitors for European cuisines — its French and Mediterranean database depth is a genuine advantage. However, Welling's global database still outperforms it even on European foods in our benchmark. For users eating primarily French or Italian food, Foodvisor's specialist coverage is worth considering.
Why is Foodvisor's portion accuracy so poor?▼
Foodvisor uses 2D pixel area scaling to estimate portions — comparing food footprint against a reference object in the image. This approach is inherently error-prone for 3D foods (thick vs. thin cuts, bowl depth, stacking). Without depth estimation, its ±31% MAPE is structurally difficult to improve without a model architecture change.