HuaZheLei
HuaZheLei
@Reagan1311 I read the paper carefully, and I find that the author adds a GAP on the tail of Context Path. Then the author just combines the up-sampled output feature...
You can just run a kNN algorithm here. For instance, in the embedding space, you could run a 3-NN search. Then, you will get the labels of the 3-NN's results....
@omoindrot We can learn embeddings through triplet loss, but we can not get the label at once. Therefore, I wonder if you can add some FC layers with a softmax...
@zijunwei Thank you very much for your reply, but I also get some problems. Just take a real annotaion file for example. > {"scores": [[3, 2, 0, 1, 1, 1,...
> mindspore里面只有配置，没有代码
@bnulihaixia Hello, could you share your caffe code with me? I am a beginner on deep learning and I have no idea about how to implement it. Thanks a lot.
@bnulihaixia Thanks for help. I will have a look.
> How many videos are you processing roughly? Does it occur if all of the paths in `video_lists` are the same video? What OS/environment are you running this under, and...
> Using the same package versions as you outlined above, on Ubuntu 22.04, I'm running all 3 different backends in a loop the same way you described above. From what...