2026-10-14T14:51:28Z https://www.scidb.cn/oai

10.57760/sciencedb.43855 2026-07-23T15:18:33Z

2026-07-23 GOOSE-OCC: Semantic Occupancy and Observation-State Annotations for Monocular Off-Road Perception doi:10.57760/sciencedb.43855 en GOOSE-OCC v1.4.0 is a derivative semantic occupancy dataset for monocular off-road perception, constructed from the publicly released GOOSE semantic data and GOOSE-DB ROS bags. It contains 3,367 pose-valid target frames, including 2,877 training frames and 490 validation frames, with a sequence-disjoint route-level split.Each target uses the front-facing CAM_WINDSHIELD_VIS camera and is associated with RGB imagery, 2D and 3D semantic annotations, LiDAR data, sensor calibration, vehicle pose, Metric3D pseudo-depth, and sparse semantic occupancy labels. The occupancy grid covers 50 m x 32 m x 8 m at an isotropic resolution of 0.25 m, resulting in a grid of 200 x 128 x 32 voxels. The taxonomy contains seven occupied semantic classes plus free space.The release also provides causal temporal indices, history RGB images, SAM3 semantic pseudo-labels with confidence maps, history Metric3D pseudo-depth, and observation-state masks. The masks distinguish known free space, visible occupied voxels, occluded observed voxels, and unknown space, enabling separate evaluation of visible perception and occluded completion with valid free-space negatives.Under the accompanying weak-supervision protocol, occupancy ground truth and LiDAR-derived observation masks are used only for validation and evaluation, not as training losses. GOOSE-OCC is distributed under the Creative Commons Attribution-ShareAlike 4.0 International license. Users must also cite the original GOOSE dataset. GOOSE-OCC; semantic occupancy; occupancy prediction; off-road robotics; monocular perception; 3D computer vision; occlusion evaluation; weak supervision; temporal perception; autonomous driving Kai Zhao Youchun Xu PUBLIC https://creativecommons.org/licenses/by-sa/4.0/ dataset Science Data Bank