{"url":"/dataset/hows","name":"HOWS","full_name":"HOWS-CL-25","description_markdown":"HOWS-CL-25 (Household Objects Within Simulation dataset for Continual Learning) is a synthetic dataset especially designed for object classification on mobile robots operating in a changing environment (like a household), where it is important to learn new, never seen objects on the fly.\r\nThis dataset can also be used for other learning use-cases, like instance segmentation or depth estimation.\r\nOr where household objects or continual learning are of interest.\r\n\r\nOur dataset contains 150,795 unique synthetic images using 25 different household categories with 925 3D models in total. For each of those categories, we generated about 6000 RGB images. In addition, we also provide a corresponding depth, segmentation, and normal image.\r\n\r\nThe dataset was created with BlenderProc [Denninger et al. (2019)], a procedural pipeline to generate images for deep learning.\r\nThis tool created a virtual room with randomly textured floors, walls, and a light source with randomly chosen light intensity and color. After that, a 3D model is placed in the resulting room. This object gets customized by randomly assigning materials, including different textures, to achieve a diverse dataset. Moreover, each object might be deformed with a random\r\ndisplacement texture.\r\nWe use 774 3D models from the ShapeNet dataset [A. X. Chang et al. (2015)] and the other models from various internet sites. Please note that we had to manually fix and filter most of the models with Blender before using them in the pipeline!\r\n\r\nFor continual learning (CL), we provide two different loading schemes:\r\n- Five sequences with five categories each\r\n- Twelve sequences with three categories in the first and two in the other sequences.\r\n\r\nIn addition to the RGB, depth, segmentation, and normal images, we also provide the calculated features of the RGB images (by ResNet50) as used in our RECALL paper.\r\nIn those two loading schemes, ten percent of the images are used for validation, where we ensure that an object instance is either in the training or the validation set, not in both. This avoids learning to recognize certain instances by heart.\r\n\r\nWe recommend using those loading schemes to compare your approach with others.\r\n\r\nFor further information and code examples, please have a look at our website: https://github.com/DLR-RM/RECALL.","description_withheld":null,"homepage":"https://zenodo.org/record/7189434","introduced_date":"2022-09-29","introduced_date_note":null,"introduced_by":{"paper":"/paper/recall-rehearsal-free-continual-learning-for","title":"RECALL: Rehearsal-free Continual Learning for Object Classification","first_author":"Markus Knauer","url":null},"license":{"name":"Creative common distributions 4.0","url":"https://creativecommons.org/licenses/by/4.0/legalcode"},"modalities":[{"name":"Images","url":"/datasets/modality/images"},{"name":"RGB-D","url":"/datasets/modality/rgb-d"}],"tasks":[{"name":"Classification","url":"/task/classification-1","datasets_with_task":"/datasets/task/classification-1"},{"name":"Continual Learning","url":"/task/continual-learning","datasets_with_task":"/datasets/task/continual-learning"},{"name":"Incremental Learning","url":"/task/incremental-learning","datasets_with_task":"/datasets/task/incremental-learning"},{"name":"Object Reconstruction","url":"/task/object-reconstruction","datasets_with_task":"/datasets/task/object-reconstruction"},{"name":"Scene Recognition","url":"/task/scene-recognition","datasets_with_task":"/datasets/task/scene-recognition"}],"languages":[{"name":"English","url":"/datasets/language/english"}],"variants":["HOWS","HOWS long"],"data_loaders":[{"repo":"https://github.com/dlr-rm/recall","url":"https://github.com/DLR-RM/RECALL#readme","frameworks":["tf"]}],"num_papers_in_archive":1,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/classification-on-hows","task":"Classification","dataset_variant":"HOWS","rows":1,"metrics":["Overall accuracy after last sequence"],"first_row_in_archive_order":{"model":"RECALL","paper":"/paper/recall-rehearsal-free-continual-learning-for","metrics":{"Overall accuracy after last sequence":"57.83"},"code_links":[{"title":"dlr-rm/recall","url":"https://github.com/dlr-rm/recall"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/classification-on-hows-long","task":"Classification","dataset_variant":"HOWS long","rows":1,"metrics":["Overall accuracy after last sequence"],"first_row_in_archive_order":{"model":"RECALL","paper":"/paper/recall-rehearsal-free-continual-learning-for","metrics":{"Overall accuracy after last sequence":"40,65"},"code_links":[{"title":"dlr-rm/recall","url":"https://github.com/dlr-rm/recall"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/recall-rehearsal-free-continual-learning-for","title":"RECALL: Rehearsal-free Continual Learning for Object Classification","date":"2022-09-29","rows_on_this_dataset":2,"code_links":1,"syntology":null}],"syntology_totals":{"read_at":"2026-09-25T09:33:49+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"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."}