AI Paper Atlas

Research map / 3D vision / 3D generation

CommonScenes: Generating Commonsense 3D Indoor Scenes with Scene Graph Diffusion

Guangyao Zhai, Evin Pinar Örnek, Shun-Cheng Wu, Yan Di, Federico Tombari, Nassir Navab, Benjamin Busam

NeurIPS 2023 · 3D vision › 3D generation · approach: Generative indoor scene synthesis · 13 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
replace DeepSDF shape decoding with relation-conditioned latent diffusion (rel2shape)Graph-to-3D style pipeline · SG-FRONT all roomsFID: 50.29 -> 45.70Improved
replace DeepSDF shape decoding with relation-conditioned latent diffusionGraph-to-3D style pipeline · SG-FRONT all roomsKID (x0.001): 7.96 -> 3.84Improved
learn layout and shape jointly instead of layout network plus text-to-shape modelCommonScenes · SG-FRONT all roomsFID: 50.58 -> 45.70Improved
add contextual graph with CLIP features and box embeddings (compare w/o shape branch to retrieval baselines)Layout branch · SG-FRONT all roomsFID: 43.51 -> 42.07Improved
add diffusion shape branch with retrieval-free generation (ablation)CommonScenes · SG-FRONT all roomsFID: 42.07 -> 45.70Hurt
use rel2shape instead of layout+txt2shapeCommonScenes · SG-FRONT dining roomFID: 64.02 -> 65.71Hurt
replace Graph-to-3D with CommonScenesCommonScenes · SG-FRONT generationsymmetrical constraint accuracy: 0.57 -> 0.60Improved
replace Graph-to-3D with CommonScenesCommonScenes · SG-FRONT generationclose by constraint accuracy: 0.74 -> 0.77Improved
replace Graph-to-3D with CommonScenesCommonScenes · SG-FRONT manipulationsymmetrical accuracy (change mode): 0.46 -> 0.59Improved
replace Graph-to-3D with CommonScenesCommonScenes · SG-FRONT diversityChamfer Distance x0.001: 1.21 -> 73.40Improved
remove CLIP contextual features, relation encoder E_r, or use concatenation instead of cross-attention (ablation)CommonScenes · SG-FRONTFID/KIDImproved
replace Graph-to-3D with CommonScenes on easy relationsCommonScenes · SG-FRONT generationleft/right accuracy: 0.98 -> 0.98No clear effect

Later papers that compared with this method

3 later accepted papers report CommonScenes as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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