I'm really beginner of semantic segementation. Therefore, I want some good source to study(Book or web site or web course…) .If you know, please share me!
", "messages": [ { "id": 2346736, "postDate": "2023-07-16T13:30:13.190Z", "content": "One good way to get started is to find a previous Kaggle competition and to read top solutions. Then try to reproduce those solutions either by using an existing notebook or coding everything from scratch. Feel free to ask things along the way, people are more than happy to help you.
\nHere is one example competition that is also segmentation: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/overview.
\nYou can find the top solution here: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/discussion/238198
(I have also written a notebook about segmentation metrics, you can check it as well: https://www.kaggle.com/code/yassinealouini/all-the-segmentation-metrics)
", "rawMarkdown": "One good way to get started is to find a previous Kaggle competition and to read top solutions. Then try to reproduce those solutions either by using an existing notebook or coding everything from scratch. Feel free to ask things along the way, people are more than happy to help you.\n\nHere is one example competition that is also segmentation: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/overview.\nYou can find the top solution here: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/discussion/238198\n\n(I have also written a notebook about segmentation metrics, you can check it as well: https://www.kaggle.com/code/yassinealouini/all-the-segmentation-metrics)", "votes": 5 }, { "id": 2346205, "postDate": "2023-07-16T05:52:29.630Z", "content": "I'm really beginner of semantic segementation. Therefore, I want some good source to study(Book or web site or web course…) .If you know, please share me!
", "rawMarkdown": "I'm really beginner of semantic segementation. Therefore, I want some good source to study(Book or web site or web course...) .If you know, please share me! \n", "votes": 3 }, { "id": 2346796, "postDate": "2023-07-16T14:25:31.967Z", "content": "It's also my first semantic segmentation experience and honnestly it's not that difficult to dial in. It's very close to classification, the main difference lies in the models, I suggest you read the U-Net paper to get started and then start implementing like @yassinealouini said.
", "rawMarkdown": "It's also my first semantic segmentation experience and honnestly it's not that difficult to dial in. It's very close to classification, the main difference lies in the models, I suggest you read the U-Net paper to get started and then start implementing like @yassinealouini said.", "votes": 2 } ], "comments": [ { "id": 2346736, "author_name": "Yassine Alouini", "author_url": "", "post_date": "2023-07-16T13:30:13.190000", "content": "One good way to get started is to find a previous Kaggle competition and to read top solutions. Then try to reproduce those solutions either by using an existing notebook or coding everything from scratch. Feel free to ask things along the way, people are more than happy to help you.
\nHere is one example competition that is also segmentation: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/overview.
\nYou can find the top solution here: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/discussion/238198
(I have also written a notebook about segmentation metrics, you can check it as well: https://www.kaggle.com/code/yassinealouini/all-the-segmentation-metrics)
", "votes": 5, "replies": [] }, { "id": 2346796, "author_name": "JEANMPIA", "author_url": "", "post_date": "2023-07-16T14:25:31.967000", "content": "It's also my first semantic segmentation experience and honnestly it's not that difficult to dial in. It's very close to classification, the main difference lies in the models, I suggest you read the U-Net paper to get started and then start implementing like @yassinealouini said.
", "votes": 2, "replies": [] } ], "raw_markdown_by_id": { "2346736": "One good way to get started is to find a previous Kaggle competition and to read top solutions. Then try to reproduce those solutions either by using an existing notebook or coding everything from scratch. Feel free to ask things along the way, people are more than happy to help you.\n\nHere is one example competition that is also segmentation: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/overview.\nYou can find the top solution here: https://www.kaggle.com/competitions/hubmap-kidney-segmentation/discussion/238198\n\n(I have also written a notebook about segmentation metrics, you can check it as well: https://www.kaggle.com/code/yassinealouini/all-the-segmentation-metrics)", "2346205": "I'm really beginner of semantic segementation. Therefore, I want some good source to study(Book or web site or web course...) .If you know, please share me! \n", "2346796": "It's also my first semantic segmentation experience and honnestly it's not that difficult to dial in. It's very close to classification, the main difference lies in the models, I suggest you read the U-Net paper to get started and then start implementing like @yassinealouini said." } }