Automatic wall detection with deep learning on top of Tensorflow framework
Автор: Coding with Vilperi
Загружено: 2023-09-27
Просмотров: 632
Описание:
This is a short demo about automatic wall detection using deeplearning. The used neural network is a U-Net network with ResNet-50 encoder implemented in top of TensorFlow network. The demo is build with React and Typescript and sent, using protobuf, to server build with c++ where the image is processed.
The implementation is based in following articles:
Lluis-Pere de las Heras et al.: Statistical segmentation and structural recognition for floor plan interpretation;
Surikov, Ilya Y. et al.: Floor Plan Recognition and Vectorization Using Combination UNet, Faster-RCNN, Statistical Component Analysis and Ramer-Douglas-Peucker
The image in demo is from CVC-FP: Database for structural floor plan analysis http://dag.cvc.uab.es/resources/floorplans/ and is licensed under Creative Commons Attribution 4.0 International License https://creativecommons.org/licenses/... with credits to
de las Heras, Llu’is-Pere and Terrades, OriolRamos and Robles, Sergi and S’anchez, Gemma: VC-FP and SGT: a new database for structural floor plan analysis and its groundtruthing tool, International Journal on Document Analysis and Recognition, 2015 .
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
