VolTex combines text-guided video segmentation with neural surface reconstruction to estimate the volume of a user-specified food item, reporting 4.46% mean volume error on the MetaFood3D benchmark.
Superpoint: Self-supervised interest point detection and description
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.GR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
VolTex: Food Volume Estimation using Text-Guided Segmentation and Neural Surface Reconstruction
VolTex combines text-guided video segmentation with neural surface reconstruction to estimate the volume of a user-specified food item, reporting 4.46% mean volume error on the MetaFood3D benchmark.