9 papers · 1 filter
LaMo: Self-Supervised Latent Motion Priors for Physical Realism in Video Generation
Bo Jiang, Depu Meng, Yihan Hu +3
Modern video generators produce visually compelling clips but still struggle with physical and motion consistency, limiting their use as reliable world simulators. Existing remedie…
VideoOdyssey: A Benchmark for Ultra-Long-Context and Omni-Modal Video Understanding
Haichen He, Jiayi Zhou, Sifeng Shang +3
Real-world long video understanding requires models to perform continuous tracking, information integration and memory retention over massive temporal spans within extreme video du…
Teaching Video Generators to Remember: Eliciting Dynamic Memory for Out-of-Sight State Evolution
Tianshuo Xu, Yichen Xie, Depu Meng +5
Video world models should maintain evolving states when evidence is unobserved, yet current generators often freeze hidden states upon interruption. This is not simply a capacity p…
SpectralSplat: Appearance-Disentangled Feed-Forward Gaussian Splatting for Driving Scenes
Quentin Herau, Tianshuo Xu, Depu Meng +5
Feed-forward 3D Gaussian Splatting methods have achieved impressive reconstruction quality for autonomous driving scenes, yet they entangle scene geometry with transient appearance…
URoPE: Universal Relative Position Embedding across Geometric Spaces
Yichen Xie, Depu Meng, Chensheng Peng +4
Relative position embedding has become a standard mechanism for encoding positional information in Transformers. However, existing formulations are typically limited to a fixed geo…
Learning to Drive is a Free Gift: Large-Scale Label-Free Autonomy Pretraining from Unposed In-The-Wild Videos
Matthew Strong, Wei-Jer Chang, Quentin Herau +4
Ego-centric driving videos available online provide an abundant source of visual data for autonomous driving, yet their lack of annotations makes it difficult to learn representati…
UniQueR: Unified Query-based Feedforward 3D Reconstruction
Chensheng Peng, Quentin Herau, Jiezhi Yang +6
We present UniQueR, a unified query-based feedforward framework for efficient and accurate 3D reconstruction from unposed images. Existing feedforward models such as DUSt3R, VGGT,…
RAYNOVA: Scale-Temporal Autoregressive World Modeling in Ray Space
Yichen Xie, Chensheng Peng, Mazen Abdelfattah +6
World foundation models aim to simulate the evolution of the real world with physically plausible behavior. Unlike prior methods that handle spatial and temporal correlations separ…
S2GO: Streaming Sparse Gaussian Occupancy Prediction
Jinhyung Park, Yihan Hu, Chensheng Peng +3
Despite the demonstrated efficiency and performance of sparse query-based representations for perception, state-of-the-art 3D occupancy prediction methods still rely on voxel-based…