CircleNet for Hip Landmark Detection

Authors: Hai Wu, Hongtao Xie, Chuanbin Liu, Zheng-Jun Zha, Jun Sun, Yongdong Zhang12370-12377

AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We construct a professional DDH dataset for the first time and evaluate our Circle Net on it. ... Our results show that the Circle Net can achieve the state-of-the-art results for landmark detection on the dataset with a large margin of 1.8 average pixels compared to current methods.
Researcher Affiliation Collaboration 1School of Information Science and Technology, University of Science and Technology of China 2Anhui Province Children s Hospital of China
Pseudocode No The paper does not contain any structured pseudocode or algorithm blocks.
Open Source Code No The dataset and source code will be publicly available.
Open Datasets No We construct a professional DDH dataset for the first time... Now, the dataset is available from authors upon reasonable request.
Dataset Splits No The paper states '7706 images are used for training and the rest 1826 images are for testing' but does not explicitly mention a validation split.
Hardware Specification Yes The Circle Net is trained using the Pytorch framework on a Ubuntu workstation equipped with an Intel i7-9700 CPU and two 11GB Nvidia Ge Force 1080Ti GPUs.
Software Dependencies No The paper mentions 'Pytorch framework' and 'Open CV' but does not provide specific version numbers for these software components.
Experiment Setup Yes During training, the mini batch size is set to 12. Adagrad optimizer is used for updating with the learning rate of 1.25e-4. The default training epoch is 30. During training, we resize the input resolution to 512 512. In our experiments, we adopt λr = 0.1 and λo = 1 as default setting.