Fusing stereo vision features with text prompts that include object class and approximate volume via a projection layer improves volume regression over vision-only baselines on public datasets.
Metafood cvpr 2024 challenge on physically informed 3d food reconstruction: Methods and results
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
PerBite applies SAM, Hunyuan3D, and Blender cleanup to paired food images for volume estimation, ranking first in the MetaFood challenge with 33.87% state-level MAPE and zero monotonicity violations.
citing papers explorer
-
Not Your Stereo-Typical Estimator: Combining Vision and Language for Volume Perception
Fusing stereo vision features with text prompts that include object class and approximate volume via a projection layer improves volume regression over vision-only baselines on public datasets.
-
PerBite: A Curated Diagnostic Workflow for Bite-Aware Food Volume Estimation
PerBite applies SAM, Hunyuan3D, and Blender cleanup to paired food images for volume estimation, ranking first in the MetaFood challenge with 33.87% state-level MAPE and zero monotonicity violations.