ECCV 2026 Oral Poppy: training-free polarization cues cut surface normal error by up to 26%
ssh4net · x · 2026-09-07
Stony Brook University's Photon Intelligence Lab announced Poppy, an ECCV 2026 long Oral paper that uses light polarization to enhance monocular surface normal estimation at test time.
- RGB normal estimators struggle on reflective, textureless, and dark surfaces; polarization encodes surface orientation independent of texture and albedo.
- Poppy is training-free: it keeps any frozen RGB backbone and optimizes per-pixel offsets to inputs and outputs plus a reflectance decomposition, supervised by a differentiable polarization rendering loss against single-shot polarization measurements.
- Across seven benchmarks and three backbone architectures (diffusion, flow, feed-forward), Poppy reduces mean angular error by 23–26% on synthetic data and 6–16% on real data without retraining. Code and paper are available.
More from Embodied
- Fourier's GR-3 humanoid climbs six floors autonomously, descends stairs untethered — chris_j_paxton · 2026-10-26
- FocusVLA routes robot camera views by predicted interaction intent across task stages — iscas-cloud · 2026-10-26
- Pandora M1: first prototype of a physical device for distributed GPU monitoring — tarantulae · 2026-10-26
- $150 modular Linux handheld Mecha Comet opens preorders — pocket SSH and Claude Code included — Paimaamu · 2026-10-26
- Apple Unveils All-New iPad mini: OLED Display, A20 Pro, Biggest Update Ever — CurieuxExplorer · 2026-10-26
- Purdue's GPS-free autonomous drone tracks runners through dense forests using on-chip LiDAR AI — KordingLab · 2026-10-26