Research
RobotPan: A 360^irc Surround-View Robotic Vision System for Embodied Perception
arXiv:2604.13476v1 Announce Type: cross Abstract: Surround-view perception is increasingly important for robotic navigation and loco-manipulation, especially in human-in-the-loop settings such as tele
arXiv:2604.13476v1 Announce Type: cross Abstract: Surround-view perception is increasingly important for robotic navigation and loco-manipulation, especially in human-in-the-loop settings such as teleoperation, data collection, and emergency takeover. However, current robotic visual interfaces are often limited to narrow forward-facing views, or, when multiple on-board cameras are available, require cumbersome manual switching that interrupts the operator's workflow. Both configurations suffer from motion-induced jitter that causes simulator sickness in head-mounted displays. We introduce a surround-view robotic vision system that combines six cameras with LiDAR to provide full 360^irc visual coverage, while meeting the geometric and real-time constraints of embodied deployment. We further present extsc{RobotPan}, a feed-forward framework that predicts metric-scaled and compact 3D Gaussians from calibrated sparse-view inputs for real-time rendering, reconstruction, and streaming. extsc{RobotPan} lifts multi-view features into a unified spherical coordinate representation and decodes Gaussians using hierarchical spherical voxel priors, allocating fine resolution near the robot and coarser resolution at larger radii to reduce computational redundancy without sacrificing fidelity. To support long sequences, our online fusion updates dynamic content while preventing unbounded growth in static regions by selectively updating appearance. Finally, we release a multi-sensor dataset tailored to 360^irc novel view synthesis and metric 3D reconstruction for robotics, covering navigation, manipulation, and locomotion on real platforms. Experiments show that extsc{RobotPan} achieves competitive quality against prior feed-forward reconstruction and view-synthesis methods while producing substantially fewer Gaussians, enabling practical real-time embodied deployment. Project website: https://robotpan.github.io/
Related
- CylinderDepth: Cylindrical Spatial Attention for Multi-View Consistent Self-Supervised Surround Depth Estimation
- GeomPrompt: Geometric Prompt Learning for RGB-D Semantic Segmentation Under Missing and Degraded Depth
- Characterizing Lidar Range-Measurement Ambiguity due to Multiple Returns
- ReaLiTy and LADS: A Unified Framework and Dataset Suite for LiDAR Adaptation Across Sensors and Adverse Weather Conditions
- Learning 3D Representations for Spatial Intelligence from Unposed Multi-View Images
Source: arXiv cs.CV | 2026-04-16