Event
Cyclops Synthesizes Sparse LiDAR Intensity into RGB Video
1 reports · 1 sources
Wei Gao et al. propose the Cyclops framework, which synthesizes sparse intensity data from non-repetitive scanning LiDAR (NRS-LiDAR) into RGB video, enabling camera-free all-weather perception. The method first uses a frozen pretrained densification module to convert sparse intensity projections into dense representations, then learns a velocity field via latent bridge matching (LBM) to transport intensity latents to the RGB distribution within a few ODE integration steps, and injects previous-frame context to suppress inter-frame flicker. The paper reports that experiments show the synthesized RGB (including that generated under near-dark conditions) enables standard RGB perception models to clearly outperform LiDAR baselines and conventional cameras across multiple lighting conditions on semantic segmentation, lane detection, and point cloud colorization tasks.
Report timeline
- Cyclops: Synthesizing Sparse LiDAR Intensity into RGB Video for Camera-Free Perception
Cyclops is a framework that synthesizes sparse Non-Repetitive Scanning LiDAR (NRS-LiDAR) intensity data into RGB video for all-weather perception tasks without a camera. Its method first uses a frozen pre-trained densification module to convert sparse intensity projections into dense representations, then learns a velocity field through Latent Bridge Matching (LBM) to transport intensity latents to the RGB distribution within a few ODE integration steps; it also injects prior-frame context via temporal attention to suppress inter-frame flickering, and models the velocity field as a policy optimized by a differentiable terminal reward.
arXiv cs.RO ·