search

dapengchen123 / video_reid

We plan to release our code in CVPR 2018 "Video Person Re-identification with Competitive Snippet-similarity Aggregation and Co-attentive Snippet Embedding"
41 stars 13 forks source link

3. can't find the LSTM in the code #3

Closed AsuradaYuci closed 5 years ago

AsuradaYuci commented 5 years ago

I can't find the LSTM in the code ,but in paper,we should use the LSTM to generate the query feature..T_T.

dapengchen123 commented 5 years ago

yes, this is not the original code, we use mean pooling instead in order to quickly debug. I plan to gradually recover the original one.

dapengchen123 commented 5 years ago

If you have time, you can help me to make this code run smoothly. I can provide the remaining code, :).

If you meet some problem, I can discuss with you.

I am now working for sensetime, so it is quite busy for me to arrange my work initially.

AsuradaYuci commented 5 years ago

😄 If you can provide the remaining code, I am glad to help you. I am a student, time is not a problem for me. Anyway, thank you very much.

dapengchen123 commented 5 years ago

@AsuradaYuci I have uploaded the lstm version.

thanks for you efforts

AsuradaYuci commented 5 years ago

@dapengchen123 I can't find the lstm version. :sob:

dapengchen123 commented 5 years ago

@AsuradaYuci it is in the self-pooling branch

AsuradaYuci commented 5 years ago

@dapengchen123 Thanks, I get it.