Course talk:CPSC532:StaRAI2020:TableDetection

Good explaination. I think it would be great to provide more information about how they made the graph for the problem in the second paper.

I'm in agreement with the other comments; it would be good to go over move of the implementation details for both papers, and include information about the architecture used within the article itself.

As with Maulik, I think it would be great if you can explain how semantic segmentation is used to achieve the 2 tasks outlined in the first section. Also, for the Figure 2 in the second paper, it'd be good if you provide some explanation to that as im not sure what is happening exactly.

In DeepDeSRT, they use semantic segmentation for parsing, what actually are the segmented objects and what are the different types?