Camera$\times$Time Factorization for Unified Multi-Camera 4D Reconstruction

Multi-camera 4D reconstruction is important for embodied perception. Synchronized robot observations naturally form a camera$\times$time structure. Cameras at the same timestamp provide complementary geometric views, while observations from the same camera capture temporal evolution. Existing methods often flatten these observations into a single sequence or process each camera independently, and thus do not explicitly model these two relations. In this paper, we first introduce a Camera$\times$Time Formulation for embodied 4D reconstruction. According to formulation, we build RoboCT4D, a unified benchmark that standardizes camera$\times$time observations and evaluation across simulated and real robot manipulation datasets. We then propose CxT4D with Camera$\times$Time Factorization. It encodes camera and temporal identities separately and alternates global attention across synchronized cameras and along individual camera streams. This design explicitly models cross-camera geometry and temporal correspondence. We further propose a Long-Sequence MV4D Strategy that reduces memory cost while preserving reconstruction accuracy. Experiments on four RoboCT4D subsets show that CxT4D improves depth and cross-camera rotation accuracy over existing methods. Our long-sequence inference reduces peak memory by 51.0% at eight timestamps and 60.3% at twelve timestamps, and enables longer inputs that otherwise run out of memory.