AI Paper Atlas

Research map / 3D vision / 3D generation

Simple and Effective Synthesis of Indoor 3D Scenes

Jing Yu Koh, Harsh Agrawal, Dhruv Batra, Richard Tucker, Austin Waters, Honglak Lee, Yinfei Yang, Jason Baldridge, Peter Anderson

AAAI 2023 · 3D vision › 3D generation · approach: Generative indoor scene synthesis · 19 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
Use the direct guidance-image-to-RGB-D GAN instead of PathdreamerIndoor novel-view synthesis on Matterport3D · Matterport3D Val-Unseen, 1 context, 1 stepFID: 27.2 -> 19.6Improved
Use the direct guidance-image-to-RGB-D GAN instead of PathdreamerIndoor novel-view synthesis on Matterport3D · Matterport3D Val-Unseen, 1 context, 1–6 stepsFID: 65.8 -> 58.0Improved
Use the direct guidance-image-to-RGB-D GAN instead of PixelSynthNovel-view synthesis · RealEstate10K val, all imagesFID: 25.5 -> 23.5Improved
Remove L1 depth reconstruction loss (ablation)Ours · Matterport3D Val-Unseen, 1 context, 1 stepFID: 19.6 -> 22.7Hurt
Add VGG perceptual loss (ablation)Ours · Matterport3D Val-Unseen, 1 contextFID: FID@1: 19.6 -> 18.2; FID@{1-6}: 58.0 -> 66.5Mixed
Add KLD loss for noise conditioning (ablation)Ours · Matterport3D Val-Unseen, 1 contextFID: FID@1: 19.6 -> 22.6; FID@{1-6}: 58.0 -> 65.2Hurt
Use one shared decoder instead of separate RGB and depth decoders (ablation)Ours · Matterport3D Val-Unseen, 1 contextFID: FID@1: 19.6 -> 24.1; FID@{1-6}: 58.0 -> 56.9Mixed
Add random guidance masking and four additional convolutional layers (ablation)Ours · Matterport3D Val-Unseen, 1 contextFID: FID@1: 24.2 -> 19.6; FID@{1-6}: 68.0 -> 58.0Improved
Replace generated depth with ground-truth depth in the GAN loss (ablation)Ours · Matterport3D Val-Unseen, 1 contextFID: FID@1: 19.6 -> 24.3; FID@{1-6}: 58.0 -> 66.1Hurt
Add synthetic trajectory augmentation using OursVLN-BERT (MURAL) · R2R Val-UnseenSuccess Rate: 67.3 -> 68.1Improved
Add Habitat renders alongside Ours synthetic trajectory augmentationVLN-BERT (MURAL) · R2R Val-UnseenSuccess Rate: 67.3 -> 68.8; +1.5%Improved
Add SimCLR image-based augmentationVLN-BERT (MURAL) · R2R Val-UnseenSuccess Rate: 67.3 -> 66.5Hurt

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