For each object detection method, the title has links pointing to the source of the experiment figures. Otherwise, figures will have an external link. I hope readers can find that easier to locate the paper rather than a reference section.

Thanks for the feedback regarding the references. As a blog, repeating information already there may actually irritate readers. So my approach is to make sure the credits are given yet make things simple. Some readers appreciate this approach and some readers may prefer a more academic approach. This is an act that not easy to balance. So I decide to use an on-demand way to find out what needs to be done rather than over thinking the process.

Sorry to hijack your discussion. Actually, my response is not targeting at your comment. But I try to address a border question of why things may not be done in some specific way. Just want to assure the feedback is not ignored.