{"url":"/dataset/stf","name":"Dense Fog","full_name":"DENSE","description_markdown":"We introduce an object detection dataset in challenging adverse weather conditions covering 12000 samples in real-world driving scenes and 1500 samples in controlled weather conditions within a fog chamber. The dataset includes different weather conditions like fog, snow, and rain and was acquired by over 10,000 km of driving in northern Europe. The driven route with cities along the road is shown on the right. In total, 100k Objekts were labeled with accurate 2D and 3D bounding boxes. The main contributions of this dataset are:\r\n- We provide a proving ground for a broad range of algorithms covering signal enhancement, domain adaptation, object detection, or multi-modal sensor fusion, focusing on the learning of robust redundancies between sensors, especially if they fail asymmetrically in different weather conditions.\r\n- The dataset was created with the initial intention to showcase methods, which learn of robust redundancies between the sensor and enable a raw data sensor fusion in case of asymmetric sensor failure induced through adverse weather effects.\r\n- In our case we departed from proposal level fusion and applied an adaptive fusion driven by measurement entropy enabling the detection also in case of unknown adverse weather effects. This method outperforms other reference fusion methods, which even drop in below single image methods.\r\n- Please check out our paper for more information.","description_withheld":null,"homepage":"https://www.uni-ulm.de/en/in/driveu/projects/dense-datasets#c811669","introduced_date":"2019-02-24","introduced_date_note":null,"introduced_by":{"paper":"/paper/seeing-through-fog-without-seeing-fog-deep","title":"Seeing Through Fog Without Seeing Fog: Deep Multimodal Sensor Fusion in Unseen Adverse Weather","first_author":"Mario Bijelic","url":null},"license":{"name":"https://github.com/princeton-computational-imaging/SeeingThroughFog/blob/master/LICENSE","url":"https://www.uni-ulm.de/en/in/driveu/projects/dense-datasets/dense-registration-form/"},"modalities":[{"name":"LiDAR","url":"/datasets/modality/lidar"}],"tasks":[{"name":"2D Object Detection","url":"/task/2d-object-detection","datasets_with_task":"/datasets/task/2d-object-detection"},{"name":"Object Detection","url":"/task/object-detection","datasets_with_task":"/datasets/task/object-detection"},{"name":"3D Object Detection","url":"/task/3d-object-detection","datasets_with_task":"/datasets/task/3d-object-detection"}],"languages":[{"name":"English","url":"/datasets/language/english"},{"name":"German","url":"/datasets/language/german"}],"variants":["Dense Fog"],"data_loaders":[],"num_papers_in_archive":20,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/2d-object-detection-on-dense-fog","task":"2D Object Detection","dataset_variant":"Dense Fog","rows":2,"metrics":["dense fog hard (AP)","light fog hard (AP)","snow/rain hard (AP)"],"first_row_in_archive_order":{"model":"HRFuser-T","paper":"/paper/hrfuser-a-multi-resolution-sensor-fusion","metrics":{"dense fog hard (AP)":"78.21","light fog hard (AP)":"86.5","snow/rain hard (AP)":"78.09"},"code_links":[{"title":"timbroed/hrfuser","url":"https://github.com/timbroed/hrfuser"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/3d-object-detection-on-stf","task":"3D Object Detection","dataset_variant":"Dense Fog","rows":1,"metrics":["mod. Car AP@.5IoU","mod. Cyclist AP@.25IoU","mod. Pedestrian AP@.25IoU","mod. mAP"],"first_row_in_archive_order":{"model":"PV-RCNN","paper":"/paper/fog-simulation-on-real-lidar-point-clouds-for","metrics":{"mod. Car AP@.5IoU":"47.38","mod. Cyclist AP@.25IoU":"27.89","mod. Pedestrian AP@.25IoU":"40.65","mod. mAP":"38.64"},"code_links":[{"title":"MartinHahner/LiDAR_fog_sim","url":"https://github.com/MartinHahner/LiDAR_fog_sim"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/hrfuser-a-multi-resolution-sensor-fusion","title":"HRFuser: A Multi-resolution Sensor Fusion Architecture for 2D Object Detection","date":"2022-06-30","rows_on_this_dataset":1,"code_links":1,"syntology":null},{"paper":"/paper/fog-simulation-on-real-lidar-point-clouds-for","title":"Fog Simulation on Real LiDAR Point Clouds for 3D Object Detection in Adverse Weather","date":"2021-08-11","rows_on_this_dataset":1,"code_links":1,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":6,"samples_ran":1,"samples_unverified":5,"pointer_only_for_licence":6,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/seeing-through-fog-without-seeing-fog-deep","title":"Seeing Through Fog Without Seeing Fog: Deep Multimodal Sensor Fusion in Unseen Adverse Weather","date":"2019-02-24","rows_on_this_dataset":1,"code_links":1,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":1,"samples_harvested":6,"samples_ran":1,"samples_unverified":5,"pointer_only_for_licence":6,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}