OmniFood8K supplies a large Chinese-food nutrition dataset and a single-image model that predicts depth then hierarchically fuses RGB and depth features in frequency space for improved nutrition estimates.
Reasoning-driven food en- ergy estimation via multimodal large language models.Nu- trients, 17(7):1128, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
OmniFood8K: Single-Image Nutrition Estimation via Hierarchical Frequency-Aligned Fusion
OmniFood8K supplies a large Chinese-food nutrition dataset and a single-image model that predicts depth then hierarchically fuses RGB and depth features in frequency space for improved nutrition estimates.