GenStream is a semantic streaming framework that replaces dense video frames with compact metadata—skeletal keypoints, camera parameters, and a static 3D background model—to enable generative reconstruction of human figures on the client side. By transmitting only structured information rather than full pixel data, it achieves over 99.9% bandwidth reduction compared to HEVC, as demonstrated on Olympic figure skating footage. The approach shifts computational load to the client and opens possibilities for volumetric avatar synthesis, multi‑view actor fusion, and personalized viewing experiences in a post‑codec era.
arXiv:2607.14935v2 Announce Type: replace Abstract: Recent advances in video understanding have spanned motion, long video, and streaming interaction, driving this field toward real-world application...
SpatialCrafter introduces a two‑stage framework for single‑image world modeling that first generates a global 3D proxy using a Point‑anchored Sparse Structure Flow module, then refines appearance with a Generative Deferred Refiner built on a video diffusion model. The method incorporates Parallel Geometry Injection and Proxy‑Aware Corruption training to integrate the proxy without disrupting the pretrained generative manifold, and it is evaluated on a newly constructed dataset of 115K scenes. Experiments demonstrate that SpatialCrafter outperforms existing approaches, reducing long‑term drift and maintaining consistency under rapid camera motion and extreme viewpoints.
arXiv:2608. 19556v1 Announce Type: cross Abstract: Streaming autoregressive diffusion models enable real-time, long-horizon video generation, but their training objectives optimize local frame prediction rather than the geometry and dynamics of a coherent world: long rollouts accumulate geometric drift and degrade into static or unnatural motion.
Streaming autoregressive diffusion models enable real-time, long-horizon video generation, but their training objectives optimize local frame prediction rather than the geometry and dynamics of a coherent world: long rollouts accumulate geometric drift and degrade into static or unnatural motion. Recent bidirectional approaches address this problem using rewards signals built upon 3D Gaussian-Splatting reconstruction.
Camera intrinsics are vital for recovering 3D structure from 2D video. However, most 3D algorithms assume fixed intrinsics throughout a video, an assumption that often fails for real-world in-the-wild videos.
arXiv:2609.14462v1 Announce Type: new Abstract: Interactive video world models must maintain broad scene context under camera motion while producing high-fidelity observations with low latency. Exist...
The paper introduces 4DStreamCtrl, a system that unifies camera motion, object trajectories, and depth into a single 3D point‑track representation, enabling joint control, depth editing, and motion transfer in a single forward pass. By mining in‑the‑wild video for 3D motion supervision and encoding it with a lightweight Geometric Motion Head, the authors train a causal streaming student that can generate arbitrarily long videos in just four denoising steps, achieving 20 FPS on a single high‑end GPU for 480p video. This approach outperforms prior camera‑only, 2D, and offline‑3D methods in motion‑control precision while maintaining temporal coherence over hundreds of frames, thereby enabling interactive 4D‑controllable streaming generation for the first time.
arXiv:2609.15032v1 Announce Type: new Abstract: Live free-viewpoint visualization of real humans is critical for immersive communication and interactive digital experiences. Existing methods either r...
BooM-VVT is a mask‑free video virtual try‑on framework that builds on a keyframe‑driven paradigm. It uses a multi‑stage training strategy with image‑level pseudo data to learn mask‑free localization, introduces Garment‑Sensitive Keyframe Sampling to capture garment appearance, and employs Frame‑Shared 3D‑RoPE for spatiotemporal correspondence. The authors also create the OmniView dataset to support diverse camera viewpoints and tasks, achieving superior temporal consistency and garment fidelity compared to existing methods.
Forge4D is a feed‑forward model that reconstructs temporally aligned 4D human representations from uncalibrated sparse‑view videos, enabling both novel view and novel time synthesis. It achieves this by jointly streaming 3D Gaussian reconstruction with dense motion prediction, using learnable state tokens for temporal consistency and a self‑supervised retargeting loss for motion prediction. Extensive experiments confirm its effectiveness on in‑domain and out‑of‑domain datasets.
BooM‑VVT is a mask‑free video virtual try‑on framework that builds on a keyframe‑driven paradigm. It introduces a multi‑stage training strategy using image‑level pseudo data to learn mask‑free localization, a garment‑sensitive keyframe sampling method to capture garment appearance, and a Frame‑Shared 3D‑RoPE module to align keyframes with target video frames for accurate garment detail transfer. The authors also release OmniView, a large‑scale multi‑view try‑on dataset, and demonstrate that BooM‑VVT outperforms existing methods in temporal consistency and garment fidelity.
arXiv:2608.21136v1 Announce Type: new Abstract: Recently, open-vocabulary zero-shot 3D scene understanding using vision foundation models has emerged as a promising alternative to data-intensive supe...