AI Paper Atlas

Research map / Video / Video generation

Aether: Geometric-Aware Unified World Modeling

Haoyi Zhu, Yifan Wang, Jianjun Zhou, Wenzheng Chang, Yang Zhou, Zizun Li, Junyi Chen, Chunhua Shen, Jiangmiao Pang, Tong He

ICCV 2025 · Video › Video generation · approach: World-model video generation · 69 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
add video depth reconstruction objective to multi-task training (ablation)Aether · action-conditioned navigationPSNR: 18.97 -> 19.70Improved
add video depth reconstruction objective (ablation)Aether · action-conditioned navigationLPIPS: 0.3074 -> 0.2659Improved
add video depth reconstruction objective (ablation)Aether · action-free visual path planningVBench weighted average: 79.59 -> 80.67Improved
post-train CogVideoX with geometric multi-task dataAether · video prediction without actionVBench weighted average: 78.51 -> 80.04Improved
add camera trajectory raymap action conditioning with post-trainingAether · action-conditioned video predictionVBench weighted average: 79.92 -> 80.71Improved
use Aether instead of CUT3R for KITTI depthAether · KITTI video depthAbs Rel: 0.118 -> 0.056Improved
use Aether instead of MonST3R-GA on BONN depthAether · BONN video depthAbs Rel: 0.067 -> 0.273Hurt
use Aether instead of CUT3R for camera pose on ScanNetAether · ScanNet camera poseATE: 0.099 -> 0.176Hurt
use Aether instead of CUT3R for camera pose on SintelAether · Sintel camera poseATE: 0.213 -> 0.189Improved
use Aether instead of DepthCrafter for depth on BONN (scale&shift)Aether · BONN video depthAbs Rel: 0.253 -> 0.308Hurt

Later papers that compared with this method

6 later accepted papers report Aether as a baseline in their results tables; 6 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