SOURCE-LINKED INTELLIGENCE

Stabilizing Camera-Controlled Novel View Synthesis at Inference Time

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Training-free, camera-controlled novel view synthesis from a single image using pre-trained video diffusion models often becomes unstable under large camera motion and long generation horizons. Existing approaches commonly combine several inference-time components, making it unclear which design choices are most important for stability. We show that the main source of stability is simple. Decomposing camera motion into small autoregressive steps limits per-step geometric distortion and reduces error accumulation. A controlled camera-step study shows that performance remains stable for small mo

Read original source ↗ Open in workspace

recordType

paper

region

Global