Skip to content

Latest commit

 

History

History
25 lines (16 loc) · 903 Bytes

File metadata and controls

25 lines (16 loc) · 903 Bytes

EdgePoint2: Compact Descriptors for Superior Efficiency and Accuracy

EdgePoint2 is a series of lightweight keypoint detection and description neural networks specifically tailored for edge computing applications. We provide 14 sub-models with various network configurations (Tiny/Small/Medium/Large/Enormous) and compact descriptor dimensions (32/48/64) to accommodate diverse usage requirements.

Install Dependencies

pip install -r requirements.txt

Usage

To run the demo using a camera, use the following command:

python demo_seq.py camera 

Alternatively, to run with a video file, use:

python demo_seq.py PATH_TO_VIDEO_FILE

You can add --model to select the sub-model from {T32, T48, S32, S48, S64, M32, M48, M64, L32, L48, L64, E32, E48, E64}. The letter indicates the model size, while the number denotes the output descriptor dimension.