We used two datasets(KITTIDC and Virtual KITTI 2.0) for training and three datasets for evaluation.

A large-scale satellite depth completion dataset for training and testing spacecraft depth completion algorithms.

This is the Aerial Dataset described in the letter "This work is described in the letter "Aerial Single-View Depth Completion with Image-Guided Uncertainty Estimation", by Lucas Teixeira, Martin R. Oswald, Marc Pollefeys, Margarita Chli, published in the IEEE Robotics and Automation Letters and presented at ICRA 2020.

The dataset was split into 6 tar-files for training and 1 for evaluation. Inside there are multiple sequences of RGB-D images compressed using HDF5. On the GitHub repository below you can find a python reader for the dataset.

The visual-inertial simulator and models that were used to create the dataset and more information are available in https://github.com/VIS4ROB-lab/aerial-depth-completion

The authors thank the creator of the 3D models. The models were downloaded from Sketchfab and all the authors are identified together with the 3D models.

The depth completion and depth prediction evaluation are related to our work published in Sparsity Invariant CNNs (THREEDV 2017). It contains over 93 thousand depth maps with corresponding raw LiDaR scans and RGB images, aligned with the "raw data" of the KITTI dataset. Given the large amount of training data, this dataset shall allow a training of complex deep learning models for the tasks of depth completion and single image depth prediction. Also, we provide manually selected images with unpublished depth maps to serve as a benchmark for those two challenging tasks.

The integration of multiple sensors is a crucial and emerging trend in the development of autonomous driving technology. The depth image obtained by stereo matching of the binocular camera is easily influenced by environment and distance. The point cloud of LiDAR has strong penetrability. However, it is much sparser than binocular images. LiDAR-stereo fusion can neutralize the advantages of the two sensors and maximize the acquisition of reliable three-dimensional information to improve the safety of automatic driving. Cross-sensor fusion is a key issue in the development of autonomous driving technology. This study proposed a real-time LiDAR-stereo depth completion network without 3D convolution to fuse point clouds and binocular images using injection guidance. At the same time, a kernel-connected spatial propagation network was utilized to refine the depth. The output of dense 3D information is more accurate for autonomous driving. Experimental results on the KITTI dataset showed that our method used real-time techniques effectively. Further, we demonstrated our solution's ability to address sensor defects and challenging environmental conditions using the p-KITTI dataset.

Bosch Industrial Depth Completion Dataset (BIDCD) is an RGBD dataset of static table top scenes with industrial objects, collected with a depth-camera from multiple Points-of-View (POV), approximately 60 for each scene.

We generated depth ground truth with a customized pipeline for removing erroneous depth values, and applied Multi-View geometry to fuse the cleaned depth frames and fill-in missing information. The fused scene mesh was back-projected to each POV, and finally a bi-lateral filter was applied to reduce the remaining holes.

For each scene we provide RGB, raw Depth, Ground-Truth Depth. Auxiliary information includes (a) workspace masks, corresponding to the footprint of workspace volume, and (b) cleaned depth, an intermediate result from the pipe-line mentioned above. For more details see our publication "BIDCD - Bosch Industrial Depth Completion Dataset".

Quantitative comparison on KITTI DC validation set.

TransCG is the first large-scale real-world dataset for transparent object depth completion and grasping, which contains 57,715 RGB-D images of 51 transparent objects and many opaque objects captured from different perspectives (~240 viewpoints) of 130 scenes under real-world settings. The samples are captured by two different types of cameras (Realsense D435 & L515). The following data is provided: The 3D model of the transparent object; The 6dpose of the transparent object in each viewpoint of each scene; The raw RGB-D image, and the ground-truth refined depth image; The mask of the transparent objects; The ground-truth surface normals of every sample.

An Open Context "predicates" dataset item. Open Context publishes structured data as granular, URL identified Web resources. This "Variables" record is part of the "Architecture and Urbanism at Seyitömer Höyük, Turkey" data publication.

Quantitative comparison on NYU Depth v2 dataset.

This Well Completion Report dataset represents an index of records from the California Department of Water Resources' (DWR) Online System for Well Completion Reports (OSWCR). This dataset is for informational purposes only. All attribute values should be verified by reviewing the original Well Completion Report. Known issues include: - Missing and duplicate records - Missing values (either missing on original Well Completion Report, or not key entered into database) - Incorrect values (e.g. incorrect Latitude, Longitude, Record Type, Planned Use, Total Completed Depth) - Limited spatial resolution: The majority of well completion reports have been spatially registered to the center of the 1x1 mile Public Land Survey System section that the well is located in.

An Open Context "predicates" dataset item. Open Context publishes structured data as granular, URL identified Web resources. This "Variables" record is part of the "Architecture and Urbanism at Seyitömer Höyük, Turkey" data publication.

ArXiv QA

(TBD) Automated ArXiv question answering via large language models Github | Homepage | Simple QA - Hugging Face Space

  Automated Question Answering with ArXiv Papers





  Latest 25 Papers

LIME: Localized Image Editing via Attention Regularization in Diffusion Models - [Arxiv] [QA]

Revisiting Depth Completion from a Stereo Matching Perspective for Cross-domain Generalization - [Arxiv] [QA]

VL-GPT: A Generative Pre-trained Transformer for Vision and… See the full description on the dataset page: https://huggingface.co/datasets/taesiri/arxiv_qa.

This Well Completion Report dataset represents an index of records from the California Department of Water Resources' (DWR) Online System for Well Completion Reports (OSWCR). This dataset is for informational purposes only. All attribute values should be verified by reviewing the original Well Completion Report. Known issues include: - Missing and duplicate records - Missing values (either missing on original Well Completion Report, or not key entered into database) - Incorrect values (e.g. incorrect Latitude, Longitude, Record Type, Planned Use, Total Completed Depth) - Limited spatial resolution: The majority of well completion reports have been spatially registered to the center of the 1x1 mile Public Land Survey System section that the well is located in.

This data release is a compilation of construction depth information for 12,383 active and inactive public-supply wells (PSWs) in California from various data sources. Construction data from multiple sources were indexed by the California State Water Resources Control Board Division of Drinking Water (DDW) primary station code (PS Code). Five different data sources were compared with the following priority order: 1, Local sources from select municipalities and water purveyors (Local); 2, Local DDW district data (DDW); 3, The United States Geological Survey (USGS) National Water Information System (NWIS); 4, The California State Water Resources Control Board Groundwater Ambient Monitoring and Assessment Groundwater Information System (SWRCB); and 5, USGS attribution of California Department of Water Resources well completion report data (WCR). For all data sources, the uppermost depth to the well's open or perforated interval was attributed as depth to top of perforations (ToP). The composite depth to bottom of well (Composite BOT) field was attributed from available construction data in the following priority order: 1, Depth to bottom of perforations (BoP); 2, Depth of completed well (Well Depth); 3; Borehole depth (Hole Depth). PSW ToPs and Composite BOTs from each of the five data sources were then compared and summary construction depths for both fields were selected for wells with multiple data sources according to the data-source priority order listed above. Case-by-case modifications to the final selected summary construction depths were made after priority order-based selection to ensure internal logical consistency (for example, ToP must not exceed Composite BOT). This data release contains eight tab-delimited text files. WellConstructionSourceData_Local.txt contains well construction-depth data, Composite BOT data-source attribution, and local agency data-source attribution for the Local data. WellConstructionSourceData_DDW.txt contains well construction-depth data and Composite BOT data-source attribution for the DDW data. WellConstructionSourceData_NWIS.txt contains well construction-depth data, Composite BOT data-source attribution, and USGS site identifiers for the NWIS data. WellConstructionSourceData_SWRCB.txt contains well construction-depth data and Composite BOT data-source attribution for the SWRCB data. WellConstructionSourceData_WCR.txt contains contains well construction depth data and Composite BOT data-source attribution for the WCR data. WellConstructionCompilation_ToP.txt contains all ToP data listed by data source. WellConstructionCompilation_BOT.txt contains all Composite BOT data listed by data source. WellConstructionCompilation_Summary.txt contains summary ToP and Composite BOT values for each well with data-source attribution for both construction fields. All construction depths are in units of feet below land surface and are reported to the nearest foot.

Global Intelligent Completion Market Report 2021 comes with the extensive industry analysis of development components, patterns, flows and sizes. The report also calculates present and past market values to forecast potential market management through the forecast period between 2021-2027. The report may be the best of what is a geographic area which expands the competitive landscape and industry perspective of the market.

The effectiveness of different components.

Abstract

The dataset was derived by the Bioregional Assessment Programme. This dataset was derived from the Gloucester geological model. You can find a link to the parent dataset in the Lineage Field in this metadata statement. The History Field in this metadata statement describes how this dataset was derived.

The data set contains the image in PNG form exported from the RMS geological model for the Gloucester Basin. There is a separate image for the legend. The image shows the eroded depth horizon layers which represents the depth structure of the eight formations found in the Basin after erosion has been applied. The image was used in ArcGIS to show the eroded depth structure for the RMS model.

Dataset History

The data set contains the image in PNG form exported from the RMS model for the Gloucester Basin. There is a separate image for the legend. The image was used in ArcGIS to show the eroded depth structure for the RMS model.Depth values were recorded in the well completion reports (GUID: 0529ae9a-4d40-460d-b52a-fb0e5f646746) for the Gloucester Basin, then compiled into text files and surfaces built in the RMS model.

Dataset Citation

Bioregional Assessment Programme (XXXX) GLO RMS Model Depth Structure Eroded v01. Bioregional Assessment Derived Dataset. Viewed 18 July 2018, http://data.bioregionalassessments.gov.au/dataset/d2d676b1-26e5-4b5f-bcf3-60ae7fe41cec.

Dataset Ancestors

• Derived From Gloucester Deep Wells Completion Reports - Geology

• Derived From GLO DEM 1sec SRTM MGA56

• Derived From Geoscience Australia, 1 second SRTM Digital Elevation Model (DEM)

• Derived From GLO Deep Well Locations and Depths of Formations V01

• Derived From GLO Geological Model Extracted Horizons Final Grid XYZ V01

Abstract

The dataset was derived by the Bioregional Assessment Programme from the Hunter RMS geological model. You can find a link to the parent dataset in the Lineage Field in this metadata statement. The History Field in this metadata statement describes how this dataset was derived.

The data set contains the ESRI grid versions of the depth structure exported from the folded and eroded RMS geological model for the Hunter subregion. RMS geomodelling was used to construct the geological model for the Hunter subregion. The model was built using data extracted from well completion reports published by mining companies and consultants which record the depth of various formations encountered during drilling works. These data were compiled into model input files (See data set: Hunter deep well completion reports - f2df86d5-6749-48c7-a445-d60067109f08) used to build the RMS model.

Nine geological formations and their depths from the surface are included covering a grid across the Basin. The geological model is based on measured depths recorded in Well completion reports published by mining companies and consultancies.

The naming convention refers to the geological age and depth (TVD ss = total vertical depth subsea reported to the Australian Height Datum) of the various formations as follows:

Regional horizon name Age (geological stage) Newcastle Coalfield Hunter Coalfield Western Coalfield Central or Southern coalfields

M600 Top Anisian Top Hawkesbury Sandstone Top Hawkesbury Sandstone Top Hawkesbury Sandstone Base Wianamatta Group

M700 Top Olenekian Base Hawkesbury Sandstone Base Hawkesbury Sandstone Base Hawkesbury Sandstone Base Hawkesbury Sandstone

P000 Top Changhsingian Base Narrabeen Group Base Narrabeen Group Base Narrabeen Group Base Narrabeen Group

P100 Upper Wuchiapingian Base Newcastle Coal Measures Base Newcastle Coal Measures Top Watts Sandstone Top Bargo Claystone

P500 Mid Capitanian Base Tomago Coal Measures Base Wittingham Coal Measures Base Illawarra Coal Measures Base Illawarra Coal Measures

P550 Top Wordian Base Mulbring Siltstone Base Mulbring Siltstone Base Berry Siltstone Base Berry Siltstone

P600 Mid Roadian Base Maitland Group Base Maitland Group Base Shoalhaven Group Base Shoalhaven Group

P700 Upper Kungurian Top Base Greta Coal Measures Top Base Greta Coal Measures

P900 Base Serpukhovian Base Seaham Formation Base Seaham Formation

with 'M' referring to Mesozoic and 'P' to Paleozoic

Coordonate system: GDA_1994_MGA_Zone_56

Purpose

The grids were used in ArcGIS maps to show the depth structure for the RMS model.

Dataset History

The data set contains the ESRI grids exported from the Hunter RMS geological model. There is a separate image for the legend. The grids were used in ArcGIS to show the depth structure for the RMS model. Depth values were recorded in the well completion reports for the Hunter subregion, then compiled into text files and surfaces built in the RMS model.

Dataset Citation

Bioregional Assessment Programme (XXXX) HUN RMS Model Depth Structure v01. Bioregional Assessment Derived Dataset. Viewed 22 June 2018, http://data.bioregionalassessments.gov.au/dataset/745ae8b3-d09a-4595-b6c9-3fe72a26e52f.

Dataset Ancestors

• Derived From HUN Deep Wells Biomarkers v01

• Derived From HUN Geological Model Inputs CSIRO 2014 to 2015 v01

• Derived From Hunter deep well completion reports

• Derived From South Sydney Deep Well Completion Reports

Well lithology records in this layer come from the Department of Environmental Conservation's Water Supply Data Composite. Managed by the Water Resources Section in the Drinking Water and Groundwater Protection Division, these database reports contain borehole lithology records submitted by Vermont licensed well drillers. Lithologic logs may contain information such as descriptions of materials encountered (e.g., sand, clay, rock, etc.), depth to bedrock, water bearing zones, etc. Data are reported in feet below ground surface. Licensed well drillers have been required to submit well completion reports on a form prepared by the Secretary of the Agency of Natural Resources for each well drilled or modified to the State since 1966. Well tags have been required since 1986. NOTE: the data contained herein are only as accurate as what was submitted to the State of Vermont by licensed well drillers; many wells have been completed but not reported, many reports have missing information, were recorded inaccurately, or poorly located geographically. Data are updated daily.Private Wells from the Well Completion Report Database can be found here: Private Wells