Number of train-masks is larger than the num of train images. But there are fewer val-masks that val images. Is it correct? Or I have error in my downloading.
", "messages": [ { "id": "614208", "postDate": "08/31/2019 08:00:46", "content": "Number of train-masks is larger than the num of train images. But there are fewer val-masks that val images. Is it correct? Or I have error in my downloading.
", "rawMarkdown": "# train (1,743,042), validation (41,620),\n# train: 1,743,042\n# train_masks: 2,125,530\n# val: 41,620\n# val_masks: 23,366\nNumber of train-masks is larger than the num of train images. But there are fewer val-masks that val images. Is it correct? Or I have error in my downloading.", "votes": null }, { "id": "614435", "postDate": "08/31/2019 13:34:24", "content": "The numbers you report seems to match the numbers reported in our page.
\n\nRegarding the annotation process please read description page, the paragraph starting with \"We annotated all boxed instances of these 350 classes on the training split that fulfill the following criteria\".
\n\nBoxes in validation seem to have slightly different statistics that in train (e.g. ratio of box to masks changes between train and val, but also does the ratio of box to image labels); which thus affect the number of resulting annotated masks. The process to select boxes to mask was identical in the train and validation set. The tool used to create the masks over the selected boxes was different, as explained here.
", "rawMarkdown": "The numbers you report seems to match the numbers [reported](https://storage.googleapis.com/openimages/web/factsfigures.html) in [our page](https://storage.googleapis.com/openimages/web/challenge2019.html).\n\nRegarding the annotation process please read [description page](https://storage.googleapis.com/openimages/web/factsfigures.html), the paragraph starting with \"We annotated all boxed instances of these 350 classes on the training split that fulfill the following criteria\".\n\nBoxes in validation seem to have slightly different statistics that in train (e.g. ratio of box to masks changes between train and val, but also does the ratio of box to image labels); which thus affect the number of resulting annotated masks. The process to select boxes to mask was identical in the train and validation set. The tool used to create the masks over the selected boxes was different, as [explained here](https://storage.googleapis.com/openimages/web/factsfigures.html).", "votes": null } ], "comments": [ { "id": 614435, "author_name": "benenson", "author_url": "", "post_date": "08/31/2019 13:34:24", "content": "The numbers you report seems to match the numbers reported in our page.
\n\nRegarding the annotation process please read description page, the paragraph starting with \"We annotated all boxed instances of these 350 classes on the training split that fulfill the following criteria\".
\n\nBoxes in validation seem to have slightly different statistics that in train (e.g. ratio of box to masks changes between train and val, but also does the ratio of box to image labels); which thus affect the number of resulting annotated masks. The process to select boxes to mask was identical in the train and validation set. The tool used to create the masks over the selected boxes was different, as explained here.
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "614208": "# train (1,743,042), validation (41,620),\n# train: 1,743,042\n# train_masks: 2,125,530\n# val: 41,620\n# val_masks: 23,366\nNumber of train-masks is larger than the num of train images. But there are fewer val-masks that val images. Is it correct? Or I have error in my downloading.", "614435": "The numbers you report seems to match the numbers [reported](https://storage.googleapis.com/openimages/web/factsfigures.html) in [our page](https://storage.googleapis.com/openimages/web/challenge2019.html).\n\nRegarding the annotation process please read [description page](https://storage.googleapis.com/openimages/web/factsfigures.html), the paragraph starting with \"We annotated all boxed instances of these 350 classes on the training split that fulfill the following criteria\".\n\nBoxes in validation seem to have slightly different statistics that in train (e.g. ratio of box to masks changes between train and val, but also does the ratio of box to image labels); which thus affect the number of resulting annotated masks. The process to select boxes to mask was identical in the train and validation set. The tool used to create the masks over the selected boxes was different, as [explained here](https://storage.googleapis.com/openimages/web/factsfigures.html)." }, "source": "meta" }