{ "created": "2023-08-22T17:33:57.287295+00:00", "files": [ { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:5fccac62d7b542ba65c057912a19e9ae", "ePIC_PID": "http://hdl.handle.net/21.11125/089a8855-ebe4-4b10-8117-68f0df7030c2", "key": "numpy_arrays.zip", "size": 13408642297, "version_id": "25f858ff-fcb7-4862-9b03-7fbe29706748" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:9b658836a036e519cfc64888b19d014e", "ePIC_PID": "http://hdl.handle.net/21.11125/b02c187a-2222-4e63-b551-34b4d739d3e3", "key": "raw_3d_station_images_test.zip", "size": 841080401, "version_id": "e8e162ad-952f-4bb0-a65a-12f52bc3a90b" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:cfedaa800a8711eca199c5576c4c0224", "ePIC_PID": "http://hdl.handle.net/21.11125/d1b6b6f3-0aee-4137-8fa2-a00f02677658", "key": "raw_3d_station_images_train_part_0.zip", "size": 4538574810, "version_id": "4738fa61-763b-4a92-b672-a14f244fae43" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:b606097a622ba0277f35804a5294c8ed", "ePIC_PID": "http://hdl.handle.net/21.11125/bb5afa82-e05c-4e78-8430-0327179f2a31", "key": "raw_3d_station_images_train_part_10.zip", "size": 4551102709, "version_id": "86cea67c-2e93-4694-8c6a-501e2804f855" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:dfbac23495174471463f1cf1a9d236fa", "ePIC_PID": "http://hdl.handle.net/21.11125/ba454f62-25d6-4091-9eaa-e60d6397bb40", "key": "raw_3d_station_images_train_part_11.zip", "size": 4542856892, "version_id": "44ed6af5-4fe6-425e-9f0e-34fcdb9c256a" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:661c4bfed2f9a0f062dacd1311944ca6", "ePIC_PID": "http://hdl.handle.net/21.11125/470cc893-a146-4f3c-aa03-adc2e50b51d2", "key": "raw_3d_station_images_train_part_12.zip", "size": 2718305021, "version_id": "b23ed817-fb83-4d0e-ae03-a461d9e1caf0" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:63809b38d3b6b20329c728fe83abacd4", "ePIC_PID": "http://hdl.handle.net/21.11125/f959013b-07fc-41b7-9fb8-9e38d8d08452", "key": "raw_3d_station_images_train_part_1.zip", "size": 4545922563, "version_id": "a5f4c29e-c18c-4699-b6ad-f48e8ace6f04" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:dd8cfb940750354556f6b01c388cef52", "ePIC_PID": "http://hdl.handle.net/21.11125/15c52847-6150-40bd-862b-84cea4678493", "key": "raw_3d_station_images_train_part_2.zip", "size": 4522250312, "version_id": "2416d15b-62bd-4c89-b872-3ab29bd95498" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:c496f3a35fcddee7462f7ad1e6772891", "ePIC_PID": "http://hdl.handle.net/21.11125/84a328f9-00d0-4903-8ec7-fbf403113341", "key": "raw_3d_station_images_train_part_3.zip", "size": 4531042064, "version_id": "aa65bc6f-c021-4486-8a8a-e5a1e75fb9dc" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:908bf508eab9d1f31754cc8964bfe86b", "ePIC_PID": "http://hdl.handle.net/21.11125/2c7a999d-0dc2-413a-bf80-73ceb31ec862", "key": "raw_3d_station_images_train_part_4.zip", "size": 4544306148, "version_id": "a5a2fecd-54d3-41f1-abf4-b471f8ce7a46" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:669db1bfa30b42a30821224c4cc66bc0", "ePIC_PID": "http://hdl.handle.net/21.11125/dbf61dee-7f8a-4d87-9bbe-4b2b6a80a9bf", "key": "raw_3d_station_images_train_part_5.zip", "size": 4547119595, "version_id": "95775ea7-6a39-4317-8694-101b673feb99" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:cd63900c2623c3b92f38d2f2c459e4fc", "ePIC_PID": "http://hdl.handle.net/21.11125/7fade359-96e3-4e11-8202-cd1f492f229e", "key": "raw_3d_station_images_train_part_6.zip", "size": 4540770752, "version_id": "8931f38a-1b90-4447-8154-7763ae827ac2" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:9da5ee8247bc281d2f6b2e8b12021b10", "ePIC_PID": "http://hdl.handle.net/21.11125/fbd97b86-43c4-4d1a-924b-4b1d84885b96", "key": "raw_3d_station_images_train_part_7.zip", "size": 4528406782, "version_id": "f9bf2aa3-77eb-4335-8933-5a6f2b7b24e8" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:028e9b5029681a32e3a01cc326280af2", "ePIC_PID": "http://hdl.handle.net/21.11125/161c5b22-f8f1-4887-8af8-c441bf384eca", "key": "raw_3d_station_images_train_part_8.zip", "size": 4539688970, "version_id": "fb5332ff-5a57-4084-90ee-2509ec3e750b" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:40868c7b70ba44b5f602eac32793a8c8", "ePIC_PID": "http://hdl.handle.net/21.11125/11a5cd9d-c934-4bbe-86f4-74175a8e7222", "key": "raw_3d_station_images_train_part_9.zip", "size": 4529528348, "version_id": "329602f0-22ba-4106-ab22-efb006e16295" }, { "bucket": "93a33faf-09b5-4df4-a433-a9593e5f8d28", "checksum": "md5:fb188510547bd26fd64fa45ce92ddc01", "ePIC_PID": "http://hdl.handle.net/21.11125/7bc018c5-7b9c-4498-ad2b-787f541b481b", "key": "seed_dataset.zip", "size": 5208787299, "version_id": "556742d0-6f8d-4720-a91f-1cde1406df3a" } ], "id": "3541bf71388946b3a3a906fef7aed491", "links": { "files": "https://b2share.fz-juelich.de/api/files/93a33faf-09b5-4df4-a433-a9593e5f8d28", "self": "https://b2share.fz-juelich.de/api/records/3541bf71388946b3a3a906fef7aed491", "versions": "https://b2share.fz-juelich.de/api/records/54fc62ec5b154963abea4f924e2d1046/versions" }, "metadata": { "$schema": "https://b2share.fz-juelich.de/api/communities/5bbb7b0d-51ed-410a-865b-69d6334a3f0e/schemas/1#/json_schema", "DOI": "https://doi.org/10.34730/3541bf71388946b3a3a906fef7aed491", "community": "5bbb7b0d-51ed-410a-865b-69d6334a3f0e", "community_specific": { "f5a12917-9ba7-4120-b016-98a1becfbe2c": { "helmholtz centre": [ "Forschungszentrum J\u00fclich" ] } }, "contact_email": "v.cherepashkin@fz-juelich.de", "creators": [ { "creator_name": "Cherepashkin, Vsevolod", "family_name": "Cherepashkin", "given_name": "Vsevolod" }, { "creator_name": "Yildiz, Erenus", "family_name": "Yildiz", "given_name": "Erenus" }, { "creator_name": "Fischbach, Andreas", "family_name": "Fischbach", "given_name": "Andreas" }, { "creator_name": "Kobbelt, Leif", "family_name": "Kobbelt", "given_name": "Leif" }, { "creator_name": "Scharr, Hanno", "family_name": "Scharr", "given_name": "Hanno" } ], "descriptions": [ { "description": "We present a new data set for 3d wheat seed reconstruction, propose a challenge \u201cWheat Seed 3d Reconstruction Challenge\u201d, and provide baseline methods [1].\n\nThe dataset consists of 2964 seeds, split into 2520 seeds for training/validation and 444 for testing. Ground truth data for the test set is not provided, however, test results can be evaluated in the \u201cWheat Seed 3d Reconstruction Challenge\u201d on https://helmholtz-data-challenges.de/.\n\nPer seed there are:\n(0) 36 images captured with a 10-degree increment in the rotation in front of the camera, ensuring a complete 360-degree view\n(1) Point cloud, reconstructed from these images with volume carving\n(2) 36 projection matrices, projecting point cloud (2) back to corresponding images (1)\n(3) Top view taken from another camera\n\nWe provide preprocessed versions of (1) and (2): (5) and (6) respectively.\n========================\nDetailed content:\n(0)\traw_3d_station_images/ -- .tif -- 1800x1000, 8 bit grayscale\n(1)\traw_point_clouds/ -- .ply -- N points, N<70000, x-y-z coordinates, ascii format\n(2)\tprojection_matrices/ -- .txt -- 36x4x4 floats\n(3)\t2d_station_images/ -- .png -- 700x700, 24 bit, RGB color\n(4)\tpreprocessed_3d_station_images/ -- .tif -- stabelized and cropped to 373x200, 8 bit, grayscale\n(5)\tpreprocessed_gt_point_clouds/ --.ply \u2013 2000 points, x-y-z coordinates, ascii format\n(6)\tgeneral_files/ -- additional files, files for competition submission\n\n(1) consists of point clouds, one per seed. (5) consists of point clouds in 36 corresponding poses per seed.\nRaw 3d station images (0) were preprocessed, stabilized and resampled, which results in (4). Raw point clouds (1) were preprocessed (resampled) such that each of them contains 2000 points, lying in the fixed set of directions /general_files/directions.csv. Preprocessed point clouds (5) are convertible to triangular mesh using indices of vertex triplets /general_files/triangles.txt. A sphere sampled with these vertices and triangles is in /general_files/fibo_msh.ply.\n\nThe test set does not include point clouds, and has 3 views per seed: 0, 120 and 240 degrees.\n\nZero-padded integers XXXX are seed indices (here depicted 0000 and 0003). Zero-padded integers YYY in rotation_YYY are rotation angles in degrees: 0, 10, .. 350 degrees.\n\nIndices of train and test sets are provided in general_files/indices_train.txt and general_files/indices_test.txt files. Volumes of raw point clouds of the train set are located in general_files/train_gt_volumes.csv.\n\nPoint cloud files .ply (1) and (5) were created with open3d python library, the header of the file is:\n\nformat ascii 1.0\nelement vertex ?????\nproperty float x\nproperty float y\nproperty float z\nend_header\n\nThese .ply are visualizable with, e.g., MeshLab (https://www.meshlab.net/) in windows.\n\nThere are 16 files in this data record: seed_dataset.zip, numpy_arrays.zip, raw_3d_station_images_train_part_X.zip with X in [0, 1, .., 12], raw_3d_station_images_test.zip\n\nThe file seed_dataset.zip contains the data, where each file is presented separately. The file numpy_arrays.zip contains numerical arrays, aggregating these separate files into multidimensional arrays readable with the python library numpy as np.load('file_name.npy'). raw_3d_station_images_*.zip contain images from turntable setup (0).\n\n\nStructure of the data after extraction of the numpy_arrays.zip into corresponding directories:\n\nnumpy_arrays\n\u251c\u2500\u2500 general_files\n\u2502 \u251c\u2500\u2500 directions.csv\n\u2502 \u251c\u2500\u2500 fibo_msh.ply\n\u2502 \u251c\u2500\u2500 indices_test.txt\n\u2502 \u251c\u2500\u2500 indices_train.txt\n\u2502 \u251c\u2500\u2500 train_gt_volumes.csv\n\u2502 \u2514\u2500\u2500 triangles.txt\n\u251c\u2500\u2500 2d_station_images_test.npy\n\u251c\u2500\u2500 2d_station_images_train.npy\n\u251c\u2500\u2500 gt_points_train.npy\n\u251c\u2500\u2500 preprocessed_3d_station_images_test.npy\n\u251c\u2500\u2500 preprocessed_3d_station_images_train.npy\n\u251c\u2500\u2500 projection_matrices_test.npy\n\u2514\u2500\u2500 projection_matrices_train.npy\n\n2d_station_images_test.npy (444, 3, 700, 700) uint8 (seed_idx_test, RGB_channel, height, width)\n2d_station_images_train.npy \t (2520, 3, 700, 700) uint8 (seed_idx_train, RGB_channel, height, width)\ngt_points_train.npy \t (2520, 36, 3, 2000) float32 (seed_idx_train, rotation_idx, x-y-z, point_idx)\npreprocessed_3d_station_images_test.npy \t(444, 3, 200, 373) uint8 (seed_idx_test, ~rotation_idx~, height, width)\npreprocessed_3d_station_images_train.npy \t(2520, 36, 200, 373) uint8 (seed_idx_train, rotation_idx, height, width)\nprojection_matrices_test.npy \t (444, 36, 4, 4) float32 (seed_idx_test, rotation_idx, row, column)\nprojection_matrices_train.npy \t (2520, 36, 4, 4) float32 (seed_idx_train, rotation_idx, row, column)\n\nrotataion_idx [0, 1, .., 35] corresponds to [0, 10, .., 350] degrees.\n~rotation_idx~~ [0, 1, 2] corresponds to [0, 120, 240] degrees.\nseed_idx_test and seed_idx_train are in following order: general_files/indices_test.txt and general_files/indices_train.txt respectively.\n\n\nStructure of the data after extraction of the seed_dataset.zip and raw_3d_station_images_*.zip into corresponding directories:\n\nseed_dataset\n\u251c\u2500\u2500 raw_3d_station_images\n\u2502 \u251c\u2500\u2500 train\n\u2502 \u2502 \u2514\u2500\u2500 0000\n\u2502 \u2502 \u251c\u2500\u2500 rotation_000.tif\n\u2502 \u2502 \u251c\u2500\u2500 rotation_010.tif\n\u2502 \u2502 \u251c\u2500\u2500 ...\n\u2502 \u2502 \u2514\u2500\u2500 rotation_350.tif\n\u2502 \u2514\u2500\u2500 test\n\u2502 \u2514\u2500\u2500 0003\n\u2502 \u251c\u2500\u2500 rotation_000.tif\n\u2502 \u251c\u2500\u2500 rotation_120.tif\n\u2502 \u2514\u2500\u2500 rotation_240.tif\n\u251c\u2500\u2500 raw_point_clouds\n\u2502 \u2514\u2500\u2500 train\n\u2502 \u2514\u2500\u2500 0000_Surface.ply\n\u251c\u2500\u2500 projection_matrices\n\u2502 \u251c\u2500\u2500 train\n\u2502 \u2502 \u2514\u2500\u2500 0000_ProjectionMatrices.txt\n\u2502 \u2514\u2500\u2500 test\n\u2502 \u2514\u2500\u2500 0003_ProjectionMatrices.txt\n\u251c\u2500\u2500 2d_station_images\n\u2502 \u251c\u2500\u2500 train\n\u2502 \u2502 \u2514\u2500\u2500 0000_2D.png\n\u2502 \u2514\u2500\u2500 test\n\u2502 \u2514\u2500\u2500 0003_2D.png\n\u251c\u2500\u2500 preprocessed_3d_station_images\n\u2502 \u251c\u2500\u2500 train\n\u2502 \u2502 \u2514\u2500\u2500 0000\n\u2502 \u2502 \u251c\u2500\u2500 rotation_000.tif\n\u2502 \u2502 \u251c\u2500\u2500 rotation_010.tif\n\u2502 \u2502 \u251c\u2500\u2500 ...\n\u2502 \u2502 \u2514\u2500\u2500 rotation_350.tif\n\u2502 \u2514\u2500\u2500 test\n\u2502 \u2514\u2500\u2500 0003\n\u2502 \u251c\u2500\u2500 rotation_000.tif\n\u2502 \u251c\u2500\u2500 rotation_120.tif\n\u2502 \u2514\u2500\u2500 rotation_240.tif\n\u251c\u2500\u2500 preprocessed_gt_point_clouds\n\u2502 \u2514\u2500\u2500 train\n\u2502 \u251c\u2500\u2500 0000\n\u2502 \u251c\u2500\u2500 rotation_000.ply\n\u2502 \u251c\u2500\u2500 rotation_010.ply\n\u2502 \u251c\u2500\u2500 ...\n\u2502 \u2514\u2500\u2500 rotation_350.ply\n\u2514\u2500\u2500 general_files\n \u251c\u2500\u2500 directions.csv\n \u251c\u2500\u2500 fibo_msh.ply\n \u251c\u2500\u2500 indices_test.txt\n \u251c\u2500\u2500 indices_train.txt\n \u251c\u2500\u2500 train_gt_volumes.csv\n \u2514\u2500\u2500 triangles.txt\n\nIt is mandatory to cite our paper when using this dataset:\n[1] V.Cherepashkin, E.Yildiz, A.Fischbach, L.Kobbelt, and H. Scharr. Deep learning based 3d reconstruction for phenotyping of wheat seeds: a dataset, challenge, and baseline method. CVPPA at ICCV 2023, Paris, 2 October 2023", "description_type": "Abstract" } ], "disciplines": [ { "classification_code": "5.1.13", "discipline_identifier": "5.1.13 \u2192 Agriculture \u2192 Plant science", "discipline_name": "5.1.13 \u2192 Agriculture \u2192 Plant science", "scheme": "b2share.legacy", "scheme_uri": "http://b2share.eudat.eu/suggest/disciplines.json" } ], "ePIC_PID": "http://hdl.handle.net/21.11125/0ca7e70c-d2b4-4fb5-8ae6-a1dbdb35b2c5", "keywords": [ { "keyword": "plant phenotyping" }, { "keyword": "plant seeds" }, { "keyword": "3d reconstruction" }, { "keyword": "deep learning" }, { "keyword": "wheat " } ], "license": { "license": "Creative Commons Attribution-NonCommercial-ShareAlike (CC-BY-NC-SA)" }, "open_access": true, "owners": [ 141 ], "publication_date": "2023-08-21T22:00:00.000Z", "publication_state": "published", "titles": [ { "title": "Deep learning based 3d reconstruction for phenotyping of wheat seeds: dataset (with raw images)" } ] }, "updated": "2023-08-22T17:35:23.515433+00:00" }