AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

Collaborative Control for Geometry-Conditioned PBR Image Generation

Shimon Vainer, Mark Boss, Mathias Parger, Konstantin Kutsy, Dante De Nigris, Ciara Rowles, Nicolas Perony, Simon Donné

ECCV 2024 · Image generation › Text-to-image diffusion · approach: Layout-to-image control · 15 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 bi-directional communication instead of one-way (ablation)Collaborative Control · Objaverse held-outCMMD (PBR): 16.44 -> 6.30Improved
use bi-directional instead of clockwise communication (ablation)Collaborative Control · Objaverse held-outCMMD (PBR): 6.78 -> 6.30Improved
use per-pixel MLP communication instead of single linear layer (ablation)Collaborative Control · Objaverse held-outCMMD (PBR): 6.30 -> 5.43No clear effect
use per-pixel linear communication instead of global attention (ablation)Collaborative Control · Objaverse held-outCMMD (PBR): 7.60 -> 6.30Improved
freeze RGB model with cross-network control instead of fine-tuning end-to-end (ablation)PBR generation · OOD prompts (50 objects)CLIPScore (Albedo): 25.04 -> 26.76Improved
freeze RGB model instead of fine-tuning, on in-distribution held-out data (ablation)PBR generation · Objaverse held-outFID (PBR): 11.41 -> 11.65No clear effect
train a dedicated PBR VAE instead of the RGB VAE on channel triplets (ablation)Collaborative Control · Objaverse held-outCMMD (PBR): 84.66 -> 6.30Improved
increase batch size 8x with 8 A100s (ablation)Collaborative Control · Objaverse held-outCMMD (PBR): 6.30 -> 2.96Improved
train at 256x256 instead of 512x512 (ablation)Collaborative Control · Objaverse held-outFID (PBR): 11.65 -> 9.82Mixed
disable text cross-attention in the PBR model (ablation)Collaborative Control with 1% training data · Objaverse held-outCMMD (PBR): 20.61 -> 6.25Improved
reduce training data from 98% to 1% (ablation)Collaborative Control (no PBR prompt attention) · Objaverse held-outCMMD (PBR): 6.30 -> 6.25No clear effect

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