Ashish Patel 🇮🇳’s Post

Day-97 Computer Vision Learning M²FCN — Multi-stage Multi-recursive-input Fully Convolutional Networks (Biomedical Image Segmentation) by Shanghai University, and Johns Hopkins University Follow me for similar post : 🇮🇳 Ashish Patel Interesting Facts : 🔸 This is a paper in 2017 ICCV with over 45 citations. 🔸 Outperforms U-Net and CUMedVision1 in EM Image Segmentation ------------------------------------------------------------------- 𝗔𝗺𝗮𝘇𝗶𝗻𝗴 𝗥𝗲𝘀𝗲𝗮𝗿𝗰𝗵 : https://lnkd.in/eQp7Zfn ------------------------------------------------------------------- 𝗜𝗠𝗣𝗢𝗥𝗧𝗔𝗡𝗖𝗘 🔸 A Holistically-nested Edge Detector (HED) network as the default sub-net, which is converted from VGG-16 network. 🔸 There are 3 main challenges in segmenting Electron Microscopy (EM) Images: 1. The variation of the membranal thickness is large, as thin as a filament to as thick as a blob. 2. The noise of EM acquisition makes the membrane contrast to be low, inducing some membranal boundaries are even invisible. 3. The presence of confounding structures, such as mitochondria and vesicles, also increases the difficulty in membrane detection. #computervision #artificialintelligence #deeplearning