fpv-iplab / rulstm

Code for the Paper: Antonino Furnari and Giovanni Maria Farinella. What Would You Expect? Anticipating Egocentric Actions with Rolling-Unrolling LSTMs and Modality Attention. International Conference on Computer Vision, 2019.
http://iplab.dmi.unict.it/rulstm
133 stars 33 forks source link

FastRCNN detector #14

Closed vmurahari3 closed 3 years ago

vmurahari3 commented 3 years ago

I was wondering if you had object recall metrics of the released FastRCNN weights? I looked at some of the detections and they didn't seem to great. Might be a bug on my end too.

antoninofurnari commented 3 years ago

Hello, unfortunately we did not compute object recall at the time. Considering that you closed the issue, is the problem solved?

vmurahari3 commented 3 years ago

Thanks for replying! I forgot to add the "background" class while visualizing the boxes and object ids were off by one as a result. The object detections look great now!