AI Paper Atlas

Research map / 3D vision / 3D generation

ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes

Chandan Yeshwanth, Yueh-Cheng Liu, Matthias Nießner, Angela Dai

ICCV 2023 · 3D vision › 3D generation · approach: Generative indoor scene synthesis · 434 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
train NVS on iPhone video instead of DSLR imagesNerfacto · ScanNet++ NVSPSNR: 25.02 -> 17.70Hurt
train NVS on iPhone video instead of DSLR imagesInstant-NGP · ScanNet++ NVSPSNR: 24.67 -> 15.18Hurt
apply a cross-scene pix2pix refiner to Nerfacto renderingsNerfacto · ScanNet++ NVS (DSLR)PSNR: 25.02 -> 25.42Improved
apply a cross-scene pix2pix refiner to Nerfacto renderingsNerfacto · ScanNet++ NVS (DSLR)LPIPS: 0.180 -> 0.156Improved
use Nerfacto (combined state-of-the-art components) instead of vanilla NeRFNVS on ScanNet++ · ScanNet++ NVS (DSLR)PSNR: 24.11 -> 25.02Improved
use KPConv instead of PointNet for 3D semantic segmentationScanNet++ semantic segmentation · ScanNet++ semantic segmentationmIoU: 0.07 -> 0.30Improved
use SoftGroup instead of PointGroup for instance segmentationScanNet++ instance segmentation · ScanNet++ instance segmentationAP50: 0.148 -> 0.237Improved

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