How to draw a curve like Figure 7

[egundogdu]

Hi, I would like to compare your method against another one and plot a curve like Figure 7 for FAUST Human dataset. Can you provide the code and necessary requirements for plotting Figure 7?

Best

[orlitany]

Will the latex file help you? I think it's saved as tikz

On Wed, Apr 25, 2018, 15:41 egundogdu [email protected] wrote:

Hi, I would like to compare your method against another one and plot a curve like Figure 7 for FAUST Human dataset. Can you provide the code and necessary requirements for plotting Figure 7?

Best

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/orlitany/DeepFunctionalMaps/issues/2, or mute the thread https://github.com/notifications/unsubscribe-auth/AEuHZuKiYWKAdSCi05sjPIHUSJ6x46sEks5tsG8QgaJpZM4TjWW7 .

[egundogdu]

What I mean is that I would like to have the code which outputs the accuracies at different geodesic distance thresholds. In other words, once I have the correspondence predictions (with respect to the template shape), I want to obtain the geodesic distances to the ground truths which I assume that you have done it for plotting the curve.

[orlitany]

I see, so basically you just need the geodesic distance map on the null shape. I can share a google drive link with the map

On Wed, Apr 25, 2018, 17:38 egundogdu [email protected] wrote:

What I mean is that I would like to have the code which outputs the accuracies at different geodesic distance thresholds. In other words, once I have the correspondence predictions (with respect to the template shape), I want to obtain the geodesic distances to the ground truths which I assume that you have done it for plotting the curve.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/orlitany/DeepFunctionalMaps/issues/2#issuecomment-384310708, or mute the thread https://github.com/notifications/unsubscribe-auth/AEuHZuSvk7kB2fLgrc-H58TmzlX6SJKHks5tsIpkgaJpZM4TjWW7 .

[mhsung]

Hi Or, Is the figure 7 for all ordered pairs (2-permutations of n) in the test set (80~99)? Actually I got a worse result than Figure 7 (dot line) when I ran the code with the given FAUST data and trained model. Would it be because the trained model is not the final version you used in the paper?

Thanks.

[lixiang-ucas]

Hi, Can you provide the code for plotting the geodesic error maps (Fig. 7)? Thanks.