How to deal with tables?

[Yashsethi24]

Firstly, thank you so much @hpanwar08 for your contributions!
I was trying to use Mask_RCNN but I observed that whenever data is a little bit spreaded like in a table, then it fails to record it. There is no bbox in that case.
Consider the following picture as a reference.

Can you help me with this?
Thank you

[hpanwar08]

Yes, it seems to a problem with the data. Data might not have a lot of examples of this kind of table.

[Yashsethi24]

Can you tell me the solution to this?
Can fine tuning the parameters help?

[hpanwar08]

Labeling data that has similar types of table might help. Then training on that data.

[Yashsethi24]

How many images do I need to train it with? Can I train your model further? If yes, how?

[hpanwar08]

Yes you can train the existing model. Process is same for retraining is same as that of new model, just add the existing model path in the config file.