Skip to content

Anomaly Detection in 3D Point Clouds using Deep Geometric Descriptors

Notifications You must be signed in to change notification settings

harsh-sutariya/anomaly_detection

Repository files navigation

Running the pretrain_main.py will pretrain the teacher network with Modelnet10 dataset. Running the student_main.py will train the student network. Running the inference.py will help in visualising the detected anomalies.

The training logs are visualised here: https://docs.google.com/document/d/1Gjj7Eo9Y-ThFrZKTDESuZt_8gTi7CBK--9xA7LLkVvA/edit?usp=sharing

This is an implementation of the paper titled "Anomaly Detection in 3D Point Clouds using Deep Geometric Descriptors" and the given task instructions.

Anomaly Visualization: https://docs.google.com/document/d/1YoxZ0qlaTDern_FNOYoJf0UBWsUolwY_yvEOnbcQt3k/edit?usp=sharing

Paper: https://arxiv.org/pdf/2202.11660

About

Anomaly Detection in 3D Point Clouds using Deep Geometric Descriptors

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages