arXiv Computer Vision
SwitchSplat: Learning Discrete Object States from Casually Captured Scenes
Read the original on arXiv Computer Vision →The Flow has not summarised this story yet — read it at arXiv Computer Vision.
Related stories
arXiv Machine Learning
Oct 15
StateCraft: Augmenting Video World Models with Abstract-State Reasoning
StateCraft introduces a new architecture that separates a persistent, updatable abstract state from the visual state in video world models. By pairing a language model with a video diffusion model, it tracks entity changes over time and conditions visual generation on this abstract state. The approach is evaluated on generative interfaces and a new benchmark, showing improved generation quality, memory usage, and controllability, while enabling direct editing of unseen states and extending to spatial 3D environments.
arXiv Computer Vision
Aug 27
4DGS-WAM: Bridging Past and Future with an Object-Centric World Action Model based on 4D Gaussian Splatting
The paper introduces 4DGS-WAM, an object‑centric world action model that uses a 4D Gaussian Splatting representation to separate dynamic objects from a static background. By predicting future actions of dynamic actors and their Gaussian splat transformations, the model can reuse previously observed static content for future state generation, reducing redundant background processing. Experiments on the KITTI‑MOT dataset demonstrate the model’s ability to perform short‑horizon prediction and past reconstruction.
arXiv Computer Vision
Oct 7
DensiTok: Making Feed-Forward 3D Gaussian Splatting See More Views Than It Is Given
arXiv:2610.07958v1 Announce Type: new Abstract: Feed-forward 3D Gaussian Splatting (3DGS) reconstructs a scene in a single forward pass, replacing per-scene optimization with a network trained across...
arXiv Computer Vision
Oct 28
BoxWorld: Let World Engines Control and Video Generators Render
BoxWorld introduces a novel approach to real‑time interactive video worlds by decoupling interaction reasoning from generative rendering. A coding agent builds a Box World Engine that applies user‑defined rules to determine interaction outcomes and exposes object states via a lightweight interface of layout boxes, persistent attributes, and dynamic status variables. A separate Box‑State Generative Renderer then autoregressively renders these evolving states into video, enabling online control over object motion, state switches, and conditional events. The system employs a three‑stage training strategy—learning object grounding from open‑domain videos, state transitions from gameplay videos, and rollout distillation on self‑generated histories—to handle real‑time state updates while maintaining visual coherence. BoxWorld outperforms end‑to‑end world models and remains competitive across multiple public benchmarks, with a new PairedStateBench benchmark for fine‑grained interaction control.
arXiv Computer Vision
Aug 25
FixAnything: 3D-Consistent Rendering Refinement via Video Generative Priors
arXiv:2608.23549v1 Announce Type: new Abstract: Rendering views using 3D scene representations such as Gaussian Splatting (3DGS), Neural Radiance Fields (NeRF), meshes, or even point clouds produces...
arXiv Computer Vision
Aug 27
3DGS-HPC: Distractor-free 3D Gaussian Splatting with Hybrid Patch-wise Classification
3DGS-HPC is a framework that improves 3D Gaussian Splatting for novel view synthesis by mitigating transient distractors such as moving objects and varying shadows. It combines a patch‑wise classification strategy that uses local spatial consistency for robust region‑level decisions with a hybrid classification metric that adaptively integrates photometric and perceptual cues. Experiments show that this approach outperforms existing methods in reducing distractor effects and enhancing 3DGS quality.