arXiv:2606. 19805v1 Announce Type: cross Abstract: Transferring the camera motion of a reference video to a freshly generated one lets creators reuse cinematic moves.
arXiv:2605.23889v2 Announce Type: replace Abstract: Online 3D reconstruction requires estimating camera pose and scene geometry under strict causal and bounded-memory constraints. Existing methods of...
arXiv:2509.04600v2 Announce Type: replace Abstract: Reconstructing global human motion from monocular video is fundamental to VR, graphics, and robotics, yet remains ill-posed due to depth ambiguity,...
CausalAlign is a new framework for diffusion‑based video generation models that tackles the problem of drifting camera ego‑motion over long horizons. It first diagnoses how ego‑motion is represented and changes during denoising, revealing a rise‑and‑fall pattern. Then it strengthens ego‑motion by aligning internal representations to a 3D foundation model in a delta‑token space that captures inter‑frame changes, improving motion accuracy, geometric consistency, and temporal smoothness while preserving visual quality.
arXiv:2606. 19676v1 Announce Type: cross Abstract: Diffusion models have achieved remarkable success in image and video generation and editing.
DefVINS is a visual‑inertial odometry pipeline tailored for deformable scenes, breaking the rigidity assumption of traditional VIO. It decomposes the odometry state into a rigid, IMU‑anchored component and a non‑rigid scene warp using an embedded deformation graph. The authors also introduce VIMandala, the first real‑world benchmark with ground‑truth camera poses for deformable VIO, and extend the synthetic Drunkard’s benchmark with inertial data, demonstrating that DefVINS outperforms both rigid and non‑rigid baselines.
Generative Cinematographer (GenCine) lifts a single image into an editable 3D scene scaffold, allowing artists to jointly author camera paths and foreground motion using local 3D motion handles. The system projects these controls into guidance maps that encode handle positions and colors across frames, enabling a pretrained video model to follow camera-relative, piecewise-rigid motion without physics simulation. Training recovers controls from real videos and synthetic ground-truth geometry, and experiments demonstrate consistent camera-relative motion, improved geometric consistency, and strong controllability across diverse real-world scenes.
arXiv:2610.29936v1 Announce Type: new Abstract: Recovering metric world-space hand motion from egocentric video supports understanding human interactions and transferring them to embodied agents. Thr...
arXiv:2608.21402v1 Announce Type: cross Abstract: World action models (WAMs) jointly denoise future video frames and robot actions, and the video prior is expected to generalize their control. Camera...
arXiv:2610.23856v1 Announce Type: new Abstract: Generated videos often lose geometric consistency under camera motion: structures warp under viewpoint change, parallax becomes inconsistent, and smoot...
Momentum introduces a two-stage pipeline for video world models that uses synchronized multi-camera rigs to represent evolving scenes. The WorldUpdate stage generates shared reference-camera videos as a memory of scene dynamics, while WorldInteract grounds target camera video generation in this memory, ensuring moving objects remain represented outside the field of view. Experiments on CARLA street scenes show a 126.2% improvement in dynamic scene consistency over static image conditioning.
The paper introduces CamTrol++, a training‑free method that stabilizes camera‑controlled novel view synthesis from a single image by decomposing large camera motions into small autoregressive steps, thereby limiting per‑step distortion and error accumulation. It also incorporates geometry‑constrained spatial attention, low‑frequency appearance anchoring, and a registration‑free warping pipeline to further enhance stability. Experiments on RealEstate10K and MegaScene demonstrate improved temporal and geometric consistency, better downstream 3D reconstruction quality, and higher generation efficiency, even for long 56‑frame sequences and under depth corruption.
SymRegFlow is a symmetry‑regularized flow‑matching framework that enables multi‑view‑consistent video generation across continuously varying camera poses without requiring ground‑truth novel‑view RGB supervision. The method geometrically warps source views into noisy anchors and uses masked dual‑anchor supervision combined with cross‑anchor denoising‑output consistency to reduce anchor‑specific errors. Experiments on Cosmos‑Drive‑Dreams and nuScenes show that SymRegFlow achieves superior video quality, achieving the lowest FVD and FVMD scores and improving FID and instance preservation compared to existing baselines.