After having spent already four (!) blog posts on the general and specific setup for my current machine learning study (data set, architecture, experimental setup, and sketch classification), it is now finally time to reveal the first results of the mapping task with respect to the shape domain. Today, we’ll focus on the transfer learning setup before looking into the multi-task learning results in one of the next posts.
In my last blog post, I gave an overview of the experiments I intend to conduct. Before, I had described the data set and the network architecture. Today, I finally report the first results. However, instead of first talking about the baseline for the mapping task (as originally intended), I will start with the classification results. The reason for this is that it makes more sense to discuss the baseline and my own transfer learning results together, since it is then much easier to compare them. Before I can talk about my own transfer learning results, I however first need to introduce the classification network on which they are based. So let’s focus focus today on the classification results that I was able to obtain with respect to the sketch data sets.
It’s been a while since my last blog post on this subject. The reason for that is simply that the neural network did not give me the results I wanted. But now it seems that I’m on a better track, so let me give you a quick update on what has changed and an overview of my next steps.
In my last blog post, I introduced my current research project: Learning to map raw input images into the shape space obtained from my prior study. Moreover, I talked a bit about the data set I used and the augmentation steps I took to increase the variety of inputs. Today, I want to share with you the network architecture which I plan to use in my experiments. So let’s get started.
This post will be the last one in my mini-series “A Similarity Space for Shapes” about joint work with Margit Scheibel. So far, I have described the overall data set, the correlation between distances and dissimilarities, and the well-shapedness of conceptual regions. Today, I will finally take a look at interpretable directions in this similarity space.