MUSES: MUlti-SEnsor Semantic perception dataset

The Multi-Sensor Semantic Perception Dataset for Driving under Uncertainty

MUSES offers 2500 multi-modal scenes, evenly distributed across various combinations of weather conditions (clear, fog, rain, and snow) and types of illumination (daytime, nighttime). Each image includes high-quality 2D pixel-level panoptic annotations and class-level and novel instance-level uncertainty annotations. Further, each adverse-condition image has a corresponding image of the same scene taken under clear-weather, daytime conditions. The annotation process for MUSES utilizes all available sensor data, allowing the annotators to also reliably label degraded image regions that are still discernible in other modalities. This results in better pixel coverage in the annotations and creates a more challenging evaluation setup. The dataset provides public benchmarks for: - Panoptic segmentation - Uncertainty-aware panoptic segmentation - Semantic segmentation - Object detection Sensor modalities included: - Frame camera (RGB) - MEMS lidar - FMCW radar - HD event camera - IMU/GNSS sensor

ImagesPoint cloudRGB-DLiDAR

벤치마크

Unsupervised Panoptic Segmentation on MUSES: MUlti-SEnsor Semantic perception dataset 결과 8개

Panoptic Segmentation on MUSES: MUlti-SEnsor Semantic perception dataset 결과 6개

Object Detection on MUSES: MUlti-SEnsor Semantic perception dataset 결과 5개

Semantic Segmentation on MUSES: MUlti-SEnsor Semantic perception dataset 결과 4개

Uncertainty-Aware Panoptic Segmentation on MUSES: MUlti-SEnsor Semantic perception dataset 결과 3개