PulseAugur / Brief
EN
LIVE 14:50:11

Brief

last 24h
[2/2] 222 sources

Multi-source AI news clustered, deduplicated, and scored 0–100 across authority, cluster strength, headline signal, and time decay.

  1. PerBite: A Curated Diagnostic Workflow for Bite-Aware Food Volume Estimation

    Researchers have developed a new workflow called PerBite for estimating food volume consumed during meals. This method uses a combination of segmentation, 3D mesh generation, and plate measurements to determine volume. PerBite achieved first place in a challenge with a low Chamfer distance and demonstrated a low MAPE for state-level volume, though consumed-volume MAPE remained higher. AI

    IMPACT Introduces a novel workflow for precise food volume estimation, potentially aiding dietary tracking and health applications.

  2. Honey, I Shrunk the Arc de Triomphe!

    Researchers have developed a new dataset called MetricScenes to address the "scale-collapse" issue in monocular geometry estimation, where distant objects are inaccurately represented. This dataset, compiled from internet photos and stereo imagery, provides metrically-grounded, in-the-wild scenes. Fine-tuning the MoGe-2 model on MetricScenes significantly improves its accuracy in estimating absolute scale for unconstrained environments. AI

    IMPACT Improves 3D scene understanding from monocular images, potentially aiding applications in robotics and augmented reality.