AI Paper Atlas

Research map / 3D vision / Depth and stereo

Neural Light Field Estimation for Street Scenes with Differentiable Virtual Object Insertion

Zian Wang, Wenzheng Chen, David Acuna, Jan Kautz, Sanja Fidler

ECCV 2022 · 3D vision › Depth and stereo · approach: Depth from light field and defocus · 34 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
add positional and peak encoding to sky encoder and decoder (ablation)sky modeling network · HDR sky reconstructionHDR MSE (x1e-2): 8.31 -> 7.49Improved
add positional and peak encoding to sky encoder and decoder (ablation)sky modeling network · HDR sky reconstructionpeak direction median angular error: 4.09° -> 3.38°Improved
add explicit peak direction and intensity to the sky vector (ablation)sky modeling network · HDR sky reconstructionHDR MSE (x1e-2): 11.01 -> 7.49Improved
add explicit peak direction and intensity to the sky vector (ablation)sky modeling network · HDR sky reconstructionpeak direction median angular error: 7.93° -> 3.38°Improved
use explicit peak direction and intensity instead of latent code with explicit azimuthsky modeling network · HDR sky reconstructionHDR MSE (x1e-2): 10.56 -> 7.49Improved
use hybrid lighting instead of baselines for sphere insertionhybrid lighting model · diffuse sphere insertionrendering MSE (x1e-3): 2.30 -> 1.79Improved
use hybrid lighting instead of baselines for sphere insertionhybrid lighting model · specular sphere insertionrendering MSE (x1e-3): 3.13 -> 2.41Improved
add six surrounding camera views as input (ablation)lighting prediction · HoliCitysun location median angular error: 22.43° -> 19.91°Improved
add six surrounding camera views as input (ablation)lighting prediction · nuScenesPSNR: 14.49 -> 17.96Improved
add six surrounding camera views as input (ablation)lighting prediction · nuScenessi-PSNR: 15.35 -> 18.45Improved
add inserted virtual objects without lighting estimation as augmentation3D object detector · nuScenes 3D detection (10%)mAP: 0.190 -> 0.201Improved
add objects rendered with estimated lighting instead of fixed dome lighting3D object detector · nuScenes 3D detection (10%)mAP: 0.201 -> 0.211Improved

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