AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

PointDiT: Pixel-Space Diffusion for Monocular Geometry Estimation

Haofei Xu, Rundi Wu, Philipp Henzler, Nikolai Kalischek, Michael Oechsle, Fabian Manhardt, Marc Pollefeys, Andreas Geiger, Federico Tombari, Michael Niemeyer

ICML 2026 · Image generation › Autoregressive and latent generation · approach: High-resolution diffusion synthesis · 0 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
predict clean x instead of velocity v (ablation)PointDiT-L · Ablation 256x256 avg 7 setsRel depth: 24.07 -> 5.54Improved
add random zero timestep with 10% probability to noise schedule (t-shift -0.8) (ablation)PointDiT-L · Ablation 256x256 avg 7 setsRel depth: 7.80 -> 6.00Improved
use DINOv3 last layer instead of linear patch embedding (ablation)PointDiT-L · Ablation 256x256 avg 7 setsRel depth: 9.64 -> 6.00Improved
use DINOv3 instead of DINOv2 last-layer features (ablation)PointDiT-L · Ablation 256x256 avg 7 setsBF1: 5.11 -> 7.24Improved
sample 4 intermediate DINOv3 layers instead of last layer (ablation)PointDiT-L · Ablation 256x256 avg 7 setsBF1: 7.24 -> 13.47Improved
use MoGe-2 4-layer features instead of DINOv3 (ablation)PointDiT-L · Ablation 256x256 avg 7 setsdelta1 point: 91.18 -> 93.35Improved
use MoGe-2 4-layer features instead of DINOv3 (ablation)PointDiT-L · Ablation 256x256 avg 7 setsBF1: 13.47 -> 11.75Hurt
add relative point loss to v-loss (ablation)PointDiT-L · Ablation 256x256 avg 7 setsBF1: 13.47 -> 13.92Improved
add relative point loss to v-loss (ablation)PointDiT-L · Ablation 256x256 avg 7 setsdelta1 depth: 95.08 -> 94.88Hurt
use generative flow matching instead of deterministic regression (ablation)PointDiT-L · Ablation 256x256 avg 7 setsBF1: 10.90 -> 13.92Improved
use patch size 16 instead of 32 at 512x512 (ablation)PointDiT-L · 512x512 6-dataset subsetBF1: 6.17 -> 10.37Improved
use patch size 16 instead of 32 at 512x512 (ablation)PointDiT-L · 512x512 6-dataset subsetRel depth: 3.48 -> 3.06Improved

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