Holoscan Sample App Data for AI-based Endoscopy Depth Estimation

This resource contains the model for depth estimation in laparoscopic videos by Ozyoruk et. al [1].

[1] Ozyoruk, K. B., Gokceler, G. I., Bobrow, T. L., Coskun, G., Incetan, K., Almalioglu, Y., ... & Turan, M. (2021). EndoSLAM dataset and an unsupervised monocular visual odometry and depth estimation approach for endoscopic videos. Medical image analysis, 71, 102058.

Model

The AI model for depth estimation in endoscopy is a fully convolutional model that expects a RGB image with size 480x270, normalized in the range [-2.0, 2.44].

Note: The provided model is in ONNX format. It will automatically be converted into a TensorRT model (.engine) the first time it is processed by a Holoscan application.

License

Refer to the license agreement for the use of the depth estimation model.