AI Paper Atlas

Research map / 3D vision / Depth and stereo

Beyond the Pixel: a Photometrically Calibrated HDR Dataset for Luminance and Color Prediction

Christophe Bolduc, Justine Giroux, Marc Hébert, Claude Demers, Jean-François Lalonde

ICCV 2023 · 3D vision › Depth and stereo · approach: Neural light field and illumination imaging · 18 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
apply input degradations (gamma, quantization, noise) to LDR input (ablation)luminance UNet · Per-pixel luminance (Laval Photometric)RMSE: 116.2 -> 121.0 (all degradations)No clear effect
narrow input FOV from 180 to 60 degrees (HDR input) (ablation)illuminance UNet · Planar illuminanceRMSE: 22.0 -> 51.8Hurt
clip HDR to LDR but keep exposure scale (ablation)illuminance UNet · Planar illuminance, 180 deg FOVRMSE: 22.0 -> 47.8Hurt
discard exposure scale from LDR input (ablation)illuminance UNet · Planar illuminance, 180 deg FOVRMSE: 47.8 -> 123.7Hurt
fine-tune on Theta jpeg inputs instead of using pretrained model directlyluminance/color networks · Theta Z1 colorRMSE: 676.4 -> 177.9Improved
fine-tune on Theta jpegilluminance network · Theta Z1 illuminanceR2: 0.314 -> 0.385Mixed
use camera jpeg instead of synthetically degraded LDR with pretrained modelpretrained networks · Theta Z1 luminanceRMSE: 130.6 -> 170.0Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy