search

pperle / gaze-data-collection

collect training and calibration data for gaze tracking
20 stars 10 forks source link

How to use this collected data in personalized calibration #2

Open DanielC-MST opened 1 year ago

DanielC-MST commented 1 year ago

Hi, I would like to check after this gaze-data collection. I got a CSV file including file_name， point_on_screen, time_till_capture, monitor_mm, and monitor_pixels. But I don't know how to use these data to calibrate in the gaze-tracking pipeline (https://github.com/pperle/gaze-tracking-pipeline).

losemankim commented 1 year ago

may be use this https://github.com/pperle/gaze-tracking

hshahid commented 1 year ago

@DanielC-MST did you manage to figure this out? I'm stuck on the same problem

hshahid commented 1 year ago

@pperle Any chance you can help out here? I have the csv file same as @DanielC-MST but not sure how to use it. :(

emrecolak55 commented 8 months ago

I'm stuck on the same. How can we use this csv in the pipeline project? @pperle

suayder commented 8 months ago

I am in the same problem, does someone found out how to fix this?

kimou6055 commented 3 months ago

anyone solved the problem? i'm facing the same issue

losemankim commented 3 months ago

eval.py 참조하세요 @kimou6055

kimou6055 commented 3 months ago

@losemankim will the eval.py do the preprocessing on my custom data or will it train another model that i can infer on for prediction?

losemankim commented 3 months ago

For preprocessing MPIIFaceGaze, download the original dataset and then run python dataset/mpii_face_gaze_preprocessing.py --input_path=./MPIIFaceGaze --output_path=./data. Or download the preprocessed dataset.

To only generate the CSV files with all filenames which gaze is not on the screen, run python dataset/mpii_face_gaze_errors.py --input_path=./MPIIFaceGaze --output_path=./data. This can be run on MPIIGaze and MPIIFaceGaze, or the CSV files can be directly downloaded for MPIIGaze and MPIIFaceGaze.

@kimou6055 if you make csv file u can use eval.py to make ckpt file

cycid commented 1 week ago

anyone solved the problem? i'm facing the same issue

Hi, did you ran the model using your calibration data?