Skip to content

tancnle/docker-darknet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

9ac68ba · May 17, 2020

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

YOLO Object Detection with Darknet

Docker Automated build Docker Cloud Build Status

A convenient way to do object detection using YOLOv3/YOLOv4 model via Docker.

Usage

Run object detection on an image

# YOLO v3
docker run --volume ${PWD}/output:/output \
           --interactive --rm tancnle/darknet-yolo:3 < interesting.png

# YOLO v4
docker run --volume ${PWD}/output:/output \
           --interactive --rm tancnle/darknet-yolo:4 < interesting.png

View the image with polygon-bounding boxes

open output/prediction.png

Example Output

Original YOLO v3 YOLO v4
original YOLO v3 YOLO v4
# YOLO v3
Loading weights from yolov3.weights...Done!
/tmp/image: Predicted in 19.697341 seconds.
pottedplant: 98%
bicycle: 89%
bicycle: 73%

# YOLO v4
Loading weights from yolov4.weights...mini_batch = 1, batch = 8, time_steps = 1, train = 0
nms_kind: greedynms (1), beta = 0.600000
nms_kind: greedynms (1), beta = 0.600000
nms_kind: greedynms (1), beta = 0.600000
Done! Loaded 162 layers from weights-file
Not compiled with OpenCV, saving to predictions.png instead

 seen 64, trained: 32032 K-images (500 Kilo-batches_64)
/tmp/image: Predicted in 22087.479000 milli-seconds.
pottedplant: 98%
bicycle: 38%
bicycle: 47%
bicycle: 53%
person: 55%
person: 67%

References

About

YOLO Image Dectection with Darknet

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published