search

skhadem / 3D-BoundingBox

PyTorch implementation for 3D Bounding Box Estimation Using Deep Learning and Geometry
MIT License
435 stars 96 forks source link

How to generate Calibration file for my own dataset? #16

Open engryasir683 opened 4 years ago

engryasir683 commented 4 years ago

i want to know the KITTI method that how did they generated calib file because when i generate it and it is totally different in syntax as compared to to that and no fruitful results. I made it by opencv chess method. kindly tell me how i can get my calib file in syntax like of the KITTIs. Thanks

nullxjx commented 4 years ago

the Calibration file is recorded from the camera when they collected the kitti dataset, so, unless you collect your own dataset, you should get the Calibration file through your own camera.that's what i am thinking currently

engryasir683 commented 4 years ago

@JiaxinTse I have my own dataset. How can I generate calibration of my camera just like KITTIs file?

nullxjx commented 4 years ago

@yasirjutt683 you might need to read the kitti datasets paper, they explained how they produce the calibration. I think when you take a photo using your own camera, you should have way to get the calibration through the camera itself, I don't know how you collect your datasets. Common datasets like MS coco or pascal voc do not have the calibration file, that's because they did not record it when they collect the pictures. you'd better go to check the original paper...By the way, currently i am also training the model on my own datasets, but the result is not good. And i am using the global calibration file, i dont konw if I can fit the model to my datasets well when I don't have calibration file for my own datasets. My datasets are just like pascal voc, with only labeled images. Do you have any ideas on how to train on your own datasets?

amineboujnah-dev commented 1 year ago

please help me with that if you already found a solution

upper127 commented 5 months ago

Now have you solved the problem?