AI Paper Atlas

Research map / 3D vision / 3D generation

Point4Cast: Streaming Dynamic Scene Reconstruction and Forecasting

Xinhang Liu, Pedro Miraldo, Suhas Lohit, Huaizu Jiang, Naoko Sawada, Yu-Wing Tai, Chi-Keung Tang, Moitreya Chatterjee

CVPR 2026 · 3D vision › 3D generation · approach: Monocular 3D object reconstruction · 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
use learned time embedding instead of sinusoidal encoding (ablation)Point4Cast (cross-attention) · PointOdyssey reconstructionAccuracy: 0.470 -> 0.437Improved
use learned time embedding instead of sinusoidal encoding (ablation)Point4Cast (cross-attention) · PointOdyssey reconstructionCompletion: 0.502 -> 0.492Improved
use FiLM conditioning instead of cross-attention for time (ablation)Point4Cast · PointOdyssey reconstructionAccuracy: 0.437 -> 0.428Improved
use FiLM conditioning instead of cross-attention for time (ablation)Point4Cast · PointOdyssey reconstructionCompletion: 0.492 -> 0.472Improved
swap VGGT backbone for CUT3R backbone (ablation)Point4Cast · PointOdyssey reconstructionAccuracy: 0.428 -> 0.410Mixed
add Point4Cast streaming spacetime representation on CUT3R backboneCUT3R · PointOdyssey reconstructionAccuracy: 0.530 -> 0.410Improved
add Point4Cast streaming spacetime representation on VGGT backboneStreamVGGT · TAPVid-3D reconstruction (zero-shot)Accuracy: 0.817 -> 0.711Improved
replace frame generation or scene-flow continuation with intrinsic time-conditioned forecastingVGGT · PointOdyssey forecasting next frameAccuracy: 0.561 -> 0.481Improved
replace frame generation or scene-flow continuation with intrinsic forecastingCUT3R scene-flow continuation · PointOdyssey forecasting next 10 framesAccuracy: 0.716 -> 0.561Improved
use Point4Cast instead of external forecasting for 10-frame horizonStreamVGGT frame generation · TAPVid-3D forecasting next 10 framesAccuracy: 1.693 -> 1.259Improved
replace forecasting with MonST3R frame generation on 10 framesPoint4Cast CUT3R · TAPVid-3D forecasting next 10 framesCompletion: 1.326 -> 1.471Hurt
add spacetime-representation scene flow, no flow supervisionMonST3R · PointOdyssey scene flow estimationEPE: 2.058 -> 1.355Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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