Research in Computer Vision (and many other fields)

Long ago I wrote an article ‘ignorance, myopia, and naivete in computer vision’ (with Tom Binford) on the state or research in around 1989. This article became quite controversial at that time. In some context I looked at the article recently. Interestingly, I can just change a few things and the paper will be relevant today also. What is interesting is that it will be relevant not only to computer vision but also to many other research fields. I think it is the basic bature of academic research that the paper talked about.

In academic research, it is easy to start considering that publications are the end result of research. The fact that publications are suppose to be a report on research (progress) not the goal is lost to most researchers. But that is not surprising because we humans often get attached to symbols rather than the real concept/object.

In any case, the following from that paper is definitely true in many research areas:

Think of solving X.
X found too complex, simplify It to x’.
x’ found too complex, simplify It to x”.
.
.
.
x”““““““““‘” IS solved, call the press and claim
that you have solved X.
Now that X has been solved, we can address Y.
( X: you name It)